Computer-implemented method for orchestrating modes of transport

Abstract

The invention relates to a computer-implemented method (100) for orchestrating modes of transport for a user's journey, comprising at least the following steps: determining (102) destinations, time context, preferences, constraints, and implicit requirements from at least one user input using at least one large-language model; determining (104) an optimal chain of modes of transport to be used, taking into account the determined destinations, time context, preferences, constraints, and implicit requirements, using a multimodal routing kernel; transmitting (106) at least one transaction command to the at least one mode of transport to be used via at least one booking and / or ticketing interface; continuously and in real time analyzing (108) traffic data, disruptions, delays, or availability;Modifying (110) the chain if, based on the analysis, the determined chain is no longer optimal, whereby the step of submitting (106) at least one transaction command with the modified chain, and the step of analyzing (108), are performed again. Procedure (100) provides efficient, personalized, and sustainable mobility integration.

Description

[0001] The invention relates to a computer-implemented method for orchestrating modes of transport.

[0002] The increasing diversity of available mobility options – private transport, local public transport, long-distance transport, sharing services, and micromobility – is leading to growing complexity in travel planning. Users today have to combine various apps, platforms, and fare systems to create an efficient and sustainable mobility chain. While existing systems offer partial functionalities such as route planning or booking, they operate in silos: they are limited in their modality, offer little adaptive response to disruptions, and only partially accommodate individual preferences.

[0003] The object of the invention is to provide a computer-implemented method for orchestrating modes of transport that provides efficient, personalized and sustainable mobility integration.

[0004] The problem is solved by the features of the independent claims. Advantageous further developments are the subject of the dependent claims and the following description.

[0005] According to one aspect, a computer-implemented method for orchestrating modes of transport for a user's journey is described, comprising at least the following steps: determining destinations, time context, preferences, constraints, and implicit requirements from at least one user input using at least one large-language model (LLM); determining an optimal chain of modes of transport to be used, taking into account the determined destinations, time context, preferences, constraints, and implicit requirements, using a multimodal routing kernel; transmitting at least one transaction command to the at least one mode of transport to be used via at least one booking and / or ticketing interface; continuously and in real time analyzing traffic data, disruptions, delays, or availability.Change the chain if, based on the analysis, the determined chain is no longer optimal, whereby the step of submitting at least one transaction command with the changed chain and the step of analyzing are performed again.

[0006] The invention describes an LLM-supported mobility agent that understands natural language input, interprets it semantically, and generates multimodal, seamless mobility chains. The agent integrates multiple modes of transport (car, public transport, micromobility, sharing services) and handles initial planning, booking, optimization, and dynamic replanning. Inputs such as: "Plan a fast and sustainable trip from Stuttgart to Munich tomorrow, but I want to cover the last kilometer by e-bike." or "I have to leave by car at 5 p.m. today, but if there's a traffic jam, I want to switch to public transport immediately." are interpreted and implemented fully automatically.

[0007] An agent is understood to be a computer program that is capable of independent and self-dynamic behavior within certain limits.

[0008] The invention provides end-to-end orchestration: The entire travel chain is controlled by the agent – ​​from planning and booking to real-time monitoring. Furthermore, it provides semantic destination interpretation: LLMs understand free speech, contextual preferences, and implicit requirements. This enables the processing of natural text or speech input. The LLM can also extract destinations, time context, preferences, constraints, and implicit requirements.

[0009] Automatic rerouting in case of disruptions will continue to be provided: The agent detects delays, cancellations, or traffic jams and immediately initiates alternative routes – including rebooking tickets or reserving other means of transport. This includes the ability to perform live analysis of traffic data, disruptions, delays, and availability.

[0010] Furthermore, cross-modal optimization is provided: All modes of transport are optimized together, based on criteria such as comfort, duration, costs, CO2 emissions, and real-time traffic conditions. This can involve the integration of various APIs (public transport, car sharing, bike sharing, navigation systems). Optimal journey chains can also be determined, taking defined criteria into account. Additionally, automated transactions, such as ticket purchase, vehicle reservation, charging, and activation of mobility services, can be provided.

[0011] The invention thus utilizes an LLM as a semantic meta-planner, generating fully automated end-to-end travel chains, optimizing across modalities, and ensuring intelligent, proactive replanning in real time. This creates a novel mobility system that, for the first time, integrates all modes of transport and all phases of use into a single AI-supported agent.

[0012] According to some examples, it is conceivable that a personalized user profile can be created using at least one learning preference model, historical mobility data and / or at least one context model that takes into account at least one time of day, at least one day of the week and / or at least one routine destination.

[0013] This allows personal preference profiles to be provided: The agent can then take into account routines, preferred means of transport, mobility budgets and sustainability goals.

[0014] According to some embodiments, it is conceivable that in the step of determining an optimal chain, at least one programming and / or application interface can be used for the at least one mode of transport to be used.

[0015] According to some examples, it is conceivable that the at least one transaction command can trigger at least one ticket purchase, at least one vehicle reservation and / or at least one charging and activation of a mobility service.

[0016] According to some embodiments, it is conceivable that the step of changing the chain can be carried out in a rule-based manner and / or supported by artificial intelligence.

[0017] This enables an autonomous replanning logic to be provided, in which rule-based and AI-supported alternative routes are determined. Re-optimization and rebooking can be carried out without user intervention.

[0018] According to another aspect, a computer program product is described, comprising instructions that, when the program is executed by a computer, cause it to perform the steps of the procedure according to the preceding description.

[0019] The advantages, effects, and further developments of the computer program product result from the advantages, effects, and further developments of the method described above. Therefore, reference is made to the preceding description in this regard. A computer program product can be understood, for example, as a data carrier on which a computer program element is stored, containing instructions executable by a computer. Alternatively or additionally, a computer program product can also be understood, for example, as a permanent or volatile data storage medium, such as flash memory or main memory, that contains the computer program element. However, this does not exclude other types of data storage media that contain the computer program element.

[0020] The invention is described below with reference to an exemplary embodiment and the accompanying drawing. It shows: Fig. 1. A flowchart of the process.

[0021] The computer-implemented method for orchestrating modes of transport for a user's journey is described below according to Fig. 1 in its entirety is designated by the reference numeral 100.

[0022] According to step 102, destinations, time context, preferences, constraints and implicit requirements are determined from at least one user input using at least one large language model.

[0023] According to step 104, an optimal chain of transport modes to be used is determined by means of a multimodal routing kernel, taking into account the identified destinations, the identified time context, the identified preferences, constraints and implicit requirements. At least one programming and / or application interface can be used for the at least one transport mode to be used.

[0024] According to a further step 106, at least one transaction command is transmitted to the at least one mode of transport to be used via at least one booking and / or ticketing interface. This at least one transaction command can trigger at least one ticket purchase, at least one vehicle reservation, and / or at least one charging and activation of a mobility service.

[0025] Furthermore, according to step 108, traffic data, disruptions, delays or availability are continuously analyzed in real time.

[0026] If the analysis reveals that the determined chain is no longer optimal, step 110 is executed, modifying the previously determined chain. This involves repeating step 106, which transmits at least one transaction command with the modified chain, and step 108, which analyzes the chain. Step 110 can be executed based on rules and / or with the support of artificial intelligence.

[0027] Furthermore, according to an optional step 112, a personalized user profile can be created using at least one learning preference model, historical mobility data and / or at least one context model that takes into account at least one time of day, at least one day of the week and / or at least one routine destination.

[0028] The example described above does not in any way limit the invention. Rather, the invention can be modified in numerous ways. All features of the invention described above can be essential to the invention, either alone or in combination.

Claims

[1] Computer-implemented method (100) for orchestrating modes of transport for a user's journey, comprising at least the following steps: a. Determine (102) destinations, time context, preferences, constraints and implicit requirements from at least one user input using at least one large language model; b. Determining (104) an optimal chain of transport modes to be used, taking into account the identified destinations, the identified time context, the identified preferences, constraints and implicit requirements, using a multimodal routing kernel; c. Transmitting (106) at least one transaction command to the at least one mode of transport to be used via at least one booking and / or ticketing interface; d. Analyzing (108) traffic data, disruptions, delays or availability continuously and in real time; e. Modify (110) the chain if, based on the analysis, the determined chain is no longer optimal, whereby the step: Submit (106) at least one transaction command, with the modified chain, and the step Analyze (108) are performed again. [2] Computer-implemented method (100) according to claim 1, characterized by , that a personalized user profile is created using at least one learning preference model, historical mobility data and / or at least one context model that takes into account at least one time of day, at least one day of the week and / or at least one routine destination (112). [3] Computer-implemented method (100) according to claim 1 or 2, characterized by , that in step: Determining (104) an optimal chain, at least one programming and / or application interface is used for the at least one mode of transport to be used. [4] Computer-implemented method (100) according to any one of the preceding claims, characterized by that at least one transaction command triggers at least one ticket purchase, at least one vehicle reservation and / or at least one charging and activation of a mobility service. [5] Computer-implemented method (100) according to any one of the preceding claims, characterized by , that the step: Change (110) of the chain is executed in a rule-based manner and / or supported by artificial intelligence. [6] Computer program product comprising instructions which, when the program is executed by a computer, cause it to perform the steps of the method (100) according to any one of claims 1 to 5.

Images