Energy consumption forecasting, analysis, and improvement proposal system for a battery-electric vehicle using a large speech model

Abstract

An energy consumption prediction analysis and improvement suggestion system for a battery electric vehicle using a large language model (LLM) comprises an estimation unit that, using an LLM, estimates the energy consumption of a battery that will be consumed on a planned journey in the battery-equipped battery electric vehicle, based on a current driving state and a previous driving history thereof, and that, using the LLM, estimates an improvement approach to bring the estimated energy consumption closer to a minimum value, and a suggestion unit that proposes the improvement approach that is estimated for the driver.

Description

BACKGROUND OF THE INVENTION 1. Field of the invention The present invention is applicable, for example, to battery electric vehicles (so-called BEVs) and so forth, which use batteries, and relates to the technical field of a system that performs a predictive analysis of the energy consumption of a battery and suggests improvements in energy consumption as needed, so that the vehicle can achieve a goal. 2. Description of the state of the art As a technology relating to this type of system, a driving range notification device for an automotive navigation system for a battery electric vehicle has been developed, in which the device displays a remaining driving range indicating how far the vehicle can travel based on the remaining charge of an in-vehicle battery and provides a notification as to whether the vehicle can return home or to a predetermined charging facility (see WO 2014 / 188652A1). SUMMARY OF THE INVENTION However, according to the state of the art described above, there is a high probability that a user or driver will not be convinced if a car navigation system only displays information about the remaining driving range and whether the destination can be reached. There are also problems with the accuracy and correctness of the information displayed in this way. In particular, without a detailed breakdown of energy consumption and an explanation of the impact of each element, it is unclear which part is most problematic. Furthermore, a technical problem arises in that a satisfactory course of action cannot be suggested if the destination cannot be reached. One object of the present invention is to provide an energy consumption prediction analysis and improvement suggestion system for a battery electric vehicle using an LLM that can predict energy consumption with a more accurate prediction accuracy and suggest improvement approaches regarding energy consumption in natural language that humans can understand. To solve the aforementioned problems, according to one aspect of the present invention, an energy consumption prediction analysis and improvement suggestion system for a battery electric vehicle using a large language model (LLM) comprises an estimation unit that, using an LLM, estimates the energy consumption of a battery on a planned route when driving to a destination in the battery-equipped battery electric vehicle, based on information relating to a driver and the driving of the battery electric vehicle, including a current driving state and a previous driving history of the battery electric vehicle and a previous driving history of battery electric vehicles of the same model as the battery electric vehicle, and which furthermore, using the LLM, estimates an improvement approach.to bring the estimated energy consumption closer to a minimum value, and a suggestion unit that proposes the improvement approach estimated for the driver. According to one aspect of the system according to the present invention, the use of LLM enables the prediction of energy consumption with a prediction accuracy that is more correct, through high-dimensional learning including text and suggesting improvement approaches to improve energy consumption in natural language that humans can understand. Such advantageous effects of the present invention will become clearer from the embodiments of the invention described below. BRIEF DESCRIPTION OF THE DRAWINGS Features, advantages and technical and industrial significance of exemplary embodiments of the invention are described below with reference to the accompanying drawings, in which the same reference numerals denote the same elements, and wherein: Fig. 1 is a block diagram representing an overall configuration of a system according to one embodiment; and Fig. 2 is a flowchart showing an example of the processing in the system according to the embodiment. DETAILED DESCRIPTION OF THE EXECUTION FORMS First, an overall configuration of an energy consumption forecast analysis and improvement suggestion system for a battery electric vehicle using a large language model (LLM) according to an embodiment (hereinafter optionally referred to simply as the “analysis and improvement suggestion system”) is described with reference to Fig. 1. The LLM used for analysis and estimation according to the present embodiment, or used for suggestions by linguistics, can be a so-called unimodal LLM or it can be a multimodal LLM. The present embodiment is designed as a system that performs an LLM analysis or AI analysis based on, for example, LLM training data, which appropriately incorporates various types of information such as driving state, previous driving history and specifications of the battery electric vehicle, various types of information relating to routes on a planned route or on maps, information relating to the previous driving history of battery electric vehicles of the same model as or similar to the battery electric vehicle, and various types of personal information relating to a user or driver of the battery electric vehicle (i.e.,Users or drivers of the battery electric vehicle and other battery electric vehicles), information about traffic laws and general knowledge relating to traffic laws, information about general knowledge, and so on. In particular, the system is designed to suggest highly accurate information about the remaining driving range and whether the destination can be reached, based on the remaining battery charge, driving style, driving maneuvers, and so on, in a way that is easily convincing to the user or driver, as described in detail below. Furthermore, by providing a detailed breakdown of energy consumption and explaining the impact of each element, the system allows the user or driver to clearly understand which aspect is most problematic. The system is also designed to suggest courses of action if the destination cannot be reached. It is important to note that such AI learning or LLM learning can utilize traditional AI learning systems, such as supervised learning, unsupervised learning, or augmented learning, as well as newer technologies like generative AI, LLMs, and so on, which have recently been implemented, are currently under development, or will be developed in the future. For example, AI learning or LLM learning can be configured here using a neural network that performs efficient learning through representational learning, transfer learning, feature selection, fine-tuning or hyperparameter tuning, ensemble learning, or similar methods. As shown in Fig. 1, the analysis and improvement suggestion system is configured according to the embodiment, including an in-vehicle unit 101 installed in a battery-electric vehicle 100 and a server unit 200. The in-vehicle unit 101 and the server unit 200 are housed in a communication network 10, such as the internet, a dedicated network, or the like. The communication network 10 also hosts a plurality or a large number of other battery-electric vehicles 100 in the same manner. Furthermore, the communication network 10 also houses an external related knowledge collection unit 301, which collects information obtained from outside the battery-electric vehicle 100 and which can be subjected to fine-tuning or hyper-tuning in the analysis and improvement suggestion system and used to generate domain knowledge (i.e.,to impart "external, related knowledge". The collection unit for external, related knowledge 301 may be provided at least partially within the server unit 200 or within a facility in which the server unit 200 is located, or may be provided within the indoor vehicle unit 101 or within the vehicle. Server Unit 200 is connected to Database 300, which stores various types of data, including data used in the Analysis and Improvement Suggestion System. Database 300 can be connected to Server Unit 200 or the In-Vehicle Unit 101 via Communication Network 10. Server Unit 200 consists of various types of computer-installed devices and various types of computer equipment that perform centralized or distributed processing; in other words, the Analysis and Improvement Suggestion System is designed as a system that performs centralized or distributed processing using the large datasets in Database 300. In Fig. 1, the battery electric vehicle 100 includes a battery 150 and is configured, for example, as a BEV. The battery electric vehicle 100 can also be a so-called hybrid electric vehicle (HEV), a plug-in HEV (PHEV), a fuel cell electric vehicle (FCEV), or the like, which uses a battery. The in-vehicle unit 101 is configured including a sensor unit 102, which includes various types of sensors arranged at predetermined positions within the vehicle; a processing unit 103, which includes a computer; a communication unit 104, which includes a modem or the like, configured to be able to communicate externally from the vehicle via the communication network 10; and an interface unit 106, which is configured to be able to exchange information with the user or driver inside the vehicle through speech and images. As one of its acquisition functions, the sensor unit 102 acquires data on the remaining charge of the battery 150 and forwards the data to the processing unit 103. The sensor unit 102 is also configured to acquire various types of information 102a relating to the current driving state of the battery electric vehicle 100 and the driver of the battery electric vehicle 100, and to forward this information to the processing unit 103 as Controller Area Network (CAN) data or the like. The processing unit 103 has a CPU that controls the sensor unit 102, the communication unit 104, the interface unit 106, the memory, and so on. It transmits various types of information relating to the driver and the journey of the battery electric vehicle 100, including the planned route of the battery electric vehicle 100, the current driving status, and the previous driving history of the battery electric vehicle 100, from the communication unit 104 to the server unit 200 as data in a predefined format. Furthermore, the interface unit 106 is configured to suggest improvement suggestions to the user or driver via the communication unit 104. These suggestions are displayed using improvement suggestion data received by the server unit 200 after processing. Under the control of the processing unit 103, the communication unit 104 transmits data collected by the battery electric vehicle 100, necessary for predictive analysis of energy consumption and improvement suggestions, via the communication network 10 to the server unit 200. Furthermore, the server unit 200 is configured to receive, via the communication network 10, the results of the predictive analysis of the energy consumption of the battery electric vehicle 100, generated using the LLM, and data relating to improvement suggestions. Interface unit 106 is configured to allow input of the destination of the battery-electric vehicle 100, conditions for selecting a planned route to the destination, and so on, via voice input or predefined operations on an image or the like. The selection of the planned route (i.e., the navigation function) can be configured so that all or part of it is executed by processing unit 103, or so that all or part of it is executed by processing unit 202 on the server unit 200 side (in other words, the in-vehicle unit 101 side primarily serves as a browser function).The interface unit 106 is further configured to be able to output the results of the predictive analysis of energy consumption and data relating to improvement suggestions, which are received from the server unit 200-page in a predetermined format, either as speech output or on an image. In Fig. 1, the server unit 200 is configured, including a communication unit 201, which includes a modem or the like, capable of communicating with each of the battery electric vehicles 100 and also with the external, related knowledge collection unit 301 via the communication network 10; a processing unit 202, which includes a computer capable of performing processing such as LLM-based energy consumption estimation processing, and so on, which will be described in detail later; and a suggestion unit 203, capable of generating suggestion data that propose improvement approaches according to the estimation results from the processing unit 202. The communication unit 201, under the control of the processing unit 202, receives data collected by the battery-electric vehicle 100 via the communication network 10. This data is necessary for predictive analysis of energy consumption and improvement suggestions. Under the control of the processing unit 202, the communication unit 201 also receives external related knowledge collected by the external related knowledge collection unit 301 via the communication network 10, as part of the data required for predictive analysis of energy consumption and improvement suggestions.The communication unit 201 is further configured to transmit to the battery electric vehicle 100, which is the subject of this analysis, the results of the predictive analysis of energy consumption and data relating to improvement suggestions processed and generated by the processing unit 202 and the suggestion unit 203, via the communication network 10 to the side of the battery electric vehicle 100. The processing unit 202 is configured to use the LLM to estimate the battery energy consumption that will be used on the planned route when the battery electric vehicle 100, which is the subject of this analysis, travels to its destination, based on information relating to the driver and the journey of the battery electric vehicle 100, including the current driving state and the previous driving history of the battery electric vehicle 100 and the previous driving history of other battery electric vehicles 100 of the same model as the battery electric vehicle 100, and also to use the LLM to estimate improvement approaches that bring the estimated energy consumption here closer to the minimum value. The suggestion unit 203 is configured to generate suggestion data for proposing improvement approaches estimated in this way to the driver or user of the battery electric vehicle 100 in a predetermined format that corresponds to the interface unit 106 provided within the battery electric vehicle 100, and to forward the data to the communication unit 201. In the present embodiment, the "suggestion unit" is thus configured to include the suggestion unit 203 on the server unit 200 side and the interface unit 106 on the in-vehicle unit 101 side, and the in-vehicle unit 101 side is primarily responsible for the browser function with respect to the suggestion function. Database 300 is configured to include a large high-speed data input / output storage device that stores various types of data received from Server Unit 200-side via Communication Network 10, in particular various types of data required for estimation processing using the LLM, data relating to the estimation results or intermediate progress generated by Processing Unit 202, suggestion data generated by Suggestion Unit 203, and so on. Next, with reference to the flowchart in Fig. 2 and in addition to the block diagram in Fig. 1, an example of the processing in the analysis and improvement suggestion system according to the present embodiment (in particular the processing which is carried out using the LLM in the processing unit 202 in the server unit 200) is described. In Fig. 2, the user first enters the "destination" for this journey using the browser function of the navigation device in the interface unit 106 of the in-vehicle unit 101. The processing unit 103 or the processing unit 202 then searches for a route to the destination via the communication network 10 (step S1). Next, processing unit 202 in server unit 200 determines whether a trip history exists for each connection of the planned route that was searched for (step S2). Some or all of these determination functions can be performed by processing unit 103 on the in-vehicle unit 101 side. Trip history data is generally stored in database 300. It should be noted that some trip history data may be stored in memory included in processing unit 103 on the in-vehicle unit 101 side, or in vehicle-internal memory provided separately from processing unit 103. If the result of the determination in step S2 is that there is no driving history (No in step S2), the "current driving state" is recorded by sensor unit 102 and processing unit 103 on the in-vehicle unit 101 side and forwarded to processing unit 202 on the server unit 200 side (step S3). The processing then proceeds to the processing, such as estimating an improvement approach using LLM in processing unit 202 or the like (step S4 and thereafter). On the other hand, if the result of the determination in step S2 is that there is a trip history (Yes in step S2), the processing unit 202 retrieves "previous trip history of BEVs of the same model as battery electric vehicle 100 for each connection" from database 300 or the like (step S5). Now, a "BEV of the same model" can include not only a BEV of exactly the same type or model as the BEV owned by the driver or user, but also a BEV with common or similar specifications that are predefined. For example, if the powertrains of both vehicles are the same or navigation numbers are the same, they can be treated as the same model here.Furthermore, slight differences between the two vehicles can be corrected by the LLM so that they can be treated as vehicles of the same model, and the corrected driving history can then be used in the processing relating to energy consumption (step S6 and thereafter). Next, processing unit 202 extracts a value corresponding to the "minimal energy consumption" from the previously obtained trip history (step S6). The processing then proceeds to a task such as estimating an improvement approach using LLM in processing unit 202 (step S4 and thereafter). Here, to estimate improvement approaches using LLM based on previous trip history (step S4 and thereafter), relationship graphs and numerical data between elements and energy consumption accumulated for each road connection are all converted to text by the LLM and then subjected to vectorization processing. The external related knowledge collection unit 301 captures external related knowledge in parallel with, before, or after the processing of steps S1 to S6 described above (step S11), and the processing of performing fine-tuning or hyper-tuning to convey domain knowledge is performed by processing unit 202 (step S12). Afterward, processing transitions to processing such as estimating an improvement approach using LLM in processing unit 202 or the like (step S4 and thereafter). Thus, the use of an LLM fine-tuned with domain knowledge related to battery electric vehicles, such as BEV domain knowledge or the like, enables the generation of improvement approaches by comparison with optimal driving settings. Next, processing unit 202 performs an estimation process to identify a difference between the driving state captured in step S3 described above and the driving state extracted in step S6 described above (step S4). The identification of the difference using the LLM is performed here based on the domain knowledge imparted in step S12 described above, in addition to the various types of data received from the in-vehicle unit 101 and the various types of data captured or extracted from the database. Next, data relating to the driving history, which was previously stored in database 300, is retrieved from database 300 as data unique to the driver of the battery electric vehicle 100 and forwarded to processing unit 202 (step S7). Next, based on the various types of data collected or generated in steps S3, S4, S6, S7, and so on, as described above, processing unit 202 generates an energy consumption improvement approach by estimation using the LLM (step S8). Here, an improvement approach that brings energy consumption closer to the minimum value is estimated by LLM as the improvement approach. The LLM, which is executed in steps S4 and S8 described above, uses various types of information as a large amount of text data, including, as appropriate, for example, the driving status, previous driving history and specifications of the battery electric vehicle 100, various types of information relating to routes on a planned route or on maps, information relating to the previous driving history of battery electric vehicles of the same model as or similar to the battery electric vehicle 100, various types of personal information relating to the user or driver of the battery electric vehicle 100 (i.e., user or driver of the battery electric vehicle 100 and other battery electric vehicles 100), information about traffic laws and general knowledge relating to traffic laws (e.g.,which side of the road to drive on, speed limits in residential areas, and so on) and information about general knowledge (e.g., is it dark at night, traffic jams are likely to occur during rush hour and during consecutive holidays, the presence of tourist attractions near the planned route, and so on), and a large amount of such information is used in verbalized form. However, it should be noted that a multimodal LLM capable of AI learning can be based not only on verbalized information but also on non-verbalized information in steps S4, S8, and so on. That is, the data used in processing steps S4, S8, and so on includes textual data but is not limited to it. In processing the steps S4, S8, and so on described above, a large amount of text data is used in this way to fine-tune and further refine the LLM. As a result, it can be applied to various types of natural language processing (NLP) tasks such as text classification, sentiment analysis, information extraction, text summarization, text generation, question answering, and so on. In step S8, the suggestion unit 203 further generates suggestion data to propose the improvement approaches generated by the LLM, and the "improvement approaches" are output as speech or images from the interface unit 106 on the inside of the vehicle unit 101. Additionally, the processing unit 202 and the suggestion unit 203 also output suggestion data for proposing "reasons" for the improvement approaches as speech or images on the interface unit 106. It should be noted that it is also preferred to use the LLM to perform the generation of suggestion data for outputting the improvement approaches, and so on, as speech or images, by the suggestion unit 203.This means that suggestions via AI language and AI images can also be made to the user or driver according to various types of natural language processing tasks such as text classification, sentiment analysis, information extraction, text summarization, text generation and question answering. Next, it is determined whether feedback is available (step S9), and if so (YES in step S9), the processing returns to step S7 and the subsequent steps are repeated, and “improvement approaches” updated through AI learning are suggested (step S8). The suggested improvements include, for example, "To reduce energy consumption on this highway, stay in your lane and drive at a constant speed of 80 km / h," "The battery charge is low, so please avoid sudden acceleration and deceleration until the next charge," "Continue driving on the road to recharge at the charging station 25 km ahead," and so on. Ultimately, the interface unit 106 on the in-vehicle unit 101 will suggest such information to the driver in the form of AI speech or AI images via voice output or image output. In the present embodiment, it is preferred to propose the reasons for the improvement approach together with the improvement approach itself from the perspective of persuading the user or driver, in other words, from the perspective of inducing the driver to follow the improvement approach. Examples of reasons for proposals here include: "(The reasons are that) the gas mileage of this car is best at 80 km / h," "(The reasons are that) repeated acceleration and deceleration from now on will deplete the charge before the next charging station is reached," and "(The reasons are that) there is no problem driving as usual to reach the charging station 25 km away," and so on. If step S9 determines that there is no feedback after processing the suggestions for improvement (No in step S9), the processing sequence ends. As described in detail above, according to the present embodiment, the previous driving history of battery electric vehicles, such as BEVs of the same type or the like, is extracted for each road connection (steps S5 and S6), the relationship between driving condition (air conditioning, vehicle speed, vehicle weight, interior temperature, and so on) and energy consumption is entered into the LLM (step S3), the difference between the driving condition with minimum energy consumption and the current settings is output by the LLM (step S4), and improvement approaches for saving energy (i.e., improvements to bring energy consumption closer to the minimum value) are suggested verbally or by speech (steps S7 and S8). Additionally, in response to user feedback (step S9), improvement approaches that take driving history more into account can be suggested from the next point in time (steps S7 and S8). This allows LLM-specific feedback mechanisms, such as enhanced learning from human feedback (RLHF), to be used to reflect driving history. This enables the collection of feedback on every element that influences energy consumption and the generation of new improvement approaches that are better tailored to the individual. Thus, the use of LLM according to the present embodiment not only enables the extraction of numerical data, as in conventional technology or prior art, but also the extraction of relevant features from textual data, such as papers and related literature, and accordingly allows energy consumption to be calculated or estimated on a higher scale. Furthermore, the use of LLM according to the present embodiment allows the user to be simultaneously provided with results, reasons, and improvement approaches in natural language or other language, thereby increasing the user's confidence in and satisfaction with the answer. Attachments The following annexes are further disclosed with regard to the embodiment described above. Annex 1 In a battery-electric vehicle equipped with a battery, an energy consumption prediction analysis and improvement suggestion system according to Annex 1 of the present invention comprises an estimation unit that, using a load-level measurement (LLM), estimates the energy consumption of the battery on a planned route when driving to a destination in the battery-electric vehicle equipped with the battery, based on information relating to a driver and the journey of the battery-electric vehicle, including a current driving state and a previous driving history of the battery-electric vehicle and a previous driving history of battery-electric vehicles of the same model as the battery-electric vehicle, and which also, using the LLM, estimates an improvement approach to bring the estimated energy consumption closer to a minimum value, and a suggestion unit.which proposes the improvement approach that is appreciated by the driver. According to the analysis and improvement suggestion system in Annex 1, the use of the LLM allows for high-dimensional learning, including text, to be used to predict energy consumption with greater accuracy. Furthermore, elements that have the greatest impact on energy consumption and potential improvements can be described in natural language that humans can understand. This also allows the collected driving history to be made available as useful information for other vehicles. Appendix 2 The analysis and improvement suggestion system of Annex 2 of the present invention is the energy consumption prediction analysis and improvement suggestion system according to Annex 1, in which the estimating unit estimates the improvement approach by identifying a difference between a current driving condition and a driving condition with minimum energy consumption in the previous driving history using the LLM. According to the analysis and improvement suggestion system in Annex 2 of the present invention, the use of the LLM to identify a difference between the current driving state and the driving state with minimum energy consumption in the previous driving history enables a relatively efficient estimation of improvement approaches to bring the energy consumption closer to the minimum value. Appendix 3 The analysis and improvement suggestion system of Annex 3 of the present invention is the energy consumption prediction analysis and improvement suggestion system according to Annex 1 or 2, in which the estimating unit acquires other external, related knowledge as information relating to the current driving state and the previous driving history, and estimates the improvement approach taking into account domain knowledge relating to the battery electric vehicle, using an LLM that is fine-tuned using the acquired external, related knowledge. According to the analysis and improvement suggestion system in Annex 3 of the present invention, improvement approaches are estimated not only based on information relating to the current driving condition and the previous driving history, but also taking into account domain knowledge with an LLM that is fine-tuned using external, related knowledge, thereby enabling a higher accuracy estimate. Appendix 4 The analysis and improvement suggestion system of Annex 4 of the present invention is the energy consumption forecast analysis and improvement suggestion system according to one of Annexes 1 to 3, in which the estimating unit acquires information relating to the driving history of the driver of the battery electric vehicle and estimates the improvement approach using the LLM in a manner that reflects the driving history by using an LLM-dedicated mechanism for feedback of the information relating to the driving history being acquired. According to the analysis and improvement suggestion system in Annex 4 of the present invention, improvement approaches are estimated by the LLM not only based on information relating to the current driving condition and the previous driving history, but also in a form that reflects the driving history, and accordingly, the suggestion of highly accurate improvement approaches suitable for the driver or user can be carried out in a more convincing manner. Appendix 5 The analysis and improvement suggestion system of Annex 5 of the present invention is the energy consumption forecast analysis and improvement suggestion system according to one of Annexes 1 to 4, in which the suggestion unit, together with the improvement approach, proposes information that indicates the reasons for estimating the improvement approach. According to the analysis and improvement suggestion system in Annex 5 of the present invention, not only improvement approaches but also the reasons for the improvement approaches are suggested, which makes it possible to propose highly precise improvement approaches in a form that is easier for the driver or user to understand. Appendix 6 In a battery-electric vehicle equipped with a battery, an analysis and improvement suggestion system according to Annex 6 of the present invention comprises estimating, using a LLM, the energy consumption of a battery consumed by driving on a planned route when driving to a destination, based on information relating to a driver of the battery-electric vehicle and the journey, including a current driving state and a previous driving history of the battery-electric vehicle and a previous driving history relating to battery-electric vehicles of the same model as the battery-electric vehicle, and also estimating, using the LLM, an improvement approach to bring the estimated energy consumption closer to a minimum value, and suggesting the improvement approach that is estimated for the driver. According to the analysis and improvement suggestion procedure in Annex 6 of the present invention, similar to the analysis and improvement suggestion system described in Annex 1, the use of the LLM enables high-dimensional learning, including text, to be used to predict energy consumption with a more accurate prediction accuracy, and elements that have the greatest impact on energy consumption and improvement approaches can also be described in natural language that humans can understand. The present invention may optionally be modified to an extent that does not contradict the core or idea of ​​the invention as can be read from the claims and the entire description, and the analysis and improvement suggestion system and method, including such modifications, are also contained in the technical idea of ​​the present invention. QUOTES INCLUDED IN THE DESCRIPTION This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature WO 2014 / 188 652 A1

[0002]

Claims

Energy consumption prediction analysis and improvement suggestion system for a battery electric vehicle using a large language model (LLM), wherein the energy consumption prediction analysis and improvement suggestion system comprises: an estimation unit that, based on information relating to a driver and the driving of the battery electric vehicle, including a current driving state and a previous driving history of the battery electric vehicle, and including the previous driving history of battery electric vehicles of the same model as the battery electric vehicle, estimates, using the LLM, the energy consumption of a battery, what energy will be consumed on a planned journey when driving to a destination in the battery-equipped battery electric vehicle, and also estimates, using the LLM, an improvement approach.to bring the estimated energy consumption closer to a minimum value; and a suggestion unit that proposes the estimated improvement approach to the driver. Energy consumption forecasting analysis and improvement suggestion system according to claim 1, wherein the estimating unit estimates the improvement approach by identifying, using the LLM, a difference between the current driving state and a driving state with minimal energy consumption in the previous driving history. Energy consumption prediction analysis and improvement suggestion system according to claim 1 or 2, wherein the estimation unit captures external, related knowledge, which excludes information relating to the current driving state and the previous driving history, and estimates the improvement approach taking into account domain knowledge relating to the battery electric vehicle using the LLM, which is fine-tuned using the captured external, related knowledge. Energy consumption forecast analysis and improvement suggestion system according to one of claims 1 to 3, wherein the estimating unit captures information relating to the personal preferences of the driver of the battery electric vehicle and estimates the improvement approach using the LLM in a manner that reflects the personal preferences by using an LLM-specific mechanism for feedback of the information relating to the captured personal preferences. Energy consumption forecast analysis and improvement suggestion system according to one of claims 1 to 4, wherein the suggestion unit together with the improvement approach proposes information that specifies the reasons for the estimate of the improvement approach. Energy consumption prediction analysis and improvement proposal procedure for a battery electric vehicle using a large language model (LLM), wherein the energy consumption prediction analysis and improvement proposal procedure comprises the following steps: Estimating the energy consumption of a battery that will be consumed on a planned trip route when driving to a destination in a battery-equipped battery electric vehicle, using the LLM, based on information relating to a driver and the driving of the battery electric vehicle, including a current driving state and a previous driving history of the battery electric vehicle, and including the previous driving history of battery electric vehicles of the same model as the battery electric vehicle, and estimating an improvement approach using the LLM.to bring the estimated energy consumption closer to a minimum value; and to suggest the estimated improvement approach to the driver.

Images