A solution for providing user guidance

Abstract

The invention relates to a method for providing user guidance in a building. The method comprises: receiving (310) user voice input from an input device (212) of a user interface unit (210), applying at least one artificial intelligence (AI) model (526) for determining (320) a user destination from the voice input and determining (330) a guidance action for guiding the user (140) to the user destination based on building related data, and generating (340) at least one guidance command to at least one output device (214a, 214b) of the user interface unit (210) to provide guidance for the user (140) in accordance with the guidance action. The invention relates also to a processing system (220), a user guidance system (200), an elevator system (100), a computer program product (525), and a computer-readable medium for providing user guidance in a building.

Description

[0001] A solution for providing user guidance

[0002] TECHNICAL FIELD

[0003] The invention concerns in general the technical field of user guidance in a building. Especially, however not exclusively, the invention concerns artificial intelligence -based a user guidance.

[0004] BACKGROUND

[0005] Finding a path to a particular destination in a building may be time-consuming and frustrating for users who are not familiar with the building, especially in multi-story buildings or in complex environments, such as shopping centers or office buildings. In the multi-story buildings, the path to the destination may further require using an elevator to move from the entry floor to another floor. Thus, there is a need for developing solutions for guiding users so that the users reach their destination in the building efficiently.

[0006] SUMMARY

[0007] The following presents a simplified summary in order to provide basic understanding of some aspects of various invention embodiments. The summary is not an extensive overview of the invention. It is neither intended to identify key or critical elements of the invention nor to delineate the scope of the invention. The following summary merely presents some concepts of the invention in a simplified form as a prelude to a more detailed description of exemplifying embodiments of the invention.

[0008] An objective of the invention is to present a method, a processing system, a user guidance system, an elevator system, a computer program product, and a computer-readable medium for providing user guidance in a building. Another objective of the invention is that the method, the processing system, the user guidance system, the elevator system, the computer program product, and the computer-readable medium for providing user guidance in a building enable an efficient solution for guiding users.

[0009] The objectives of the invention are reached by a method, a processing system, a user guidance system, an elevator system, a computer program product, and a computer-readable medium as defined by the respective independent claims. According to a first aspect, a method for providing user guidance in a building is provided, wherein the method comprises: receiving user voice input from an input device of a user interface unit, applying at least one artificial intelligence (Al) model for: determining a user destination from the voice input; and determining a guidance action for guiding the user to the user destination based on building related data, and generating at least one guidance command to at least one output device of the user interface unit to provide guidance for the user in accordance with the guidance action.

[0010] The method may further comprise applying the at least one Al model for postprocessing the determined guidance action for detecting and filtering out hallucinations in the determined guidance action, wherein the post-processing of the determined guidance action may comprise at least one of the following: checking the determined guidance action against the building related data, performing a consistency check, performing a redundancy and repetition check, performing a contextual verification, performing a temporal verification, comparing guidance actions determined by multiple Al models, performing language and grammar correction.

[0011] The method may further comprise in response to the detection of the hallucinations in the determined guidance action generating a command to the at least one output device to provide a request for the user for providing further voice input comprising further indication of the user destination.

[0012] Alternatively or in addition, the method may further comprise generating at least one elevator command to an elevator control unit of an elevator system to control at least one operation of the elevator system in accordance with the guidance action to assist the user to reach the user destination.

[0013] Alternatively or in addition, the method may further comprise applying the at least one Al model for pre-processing the received user voice input for detecting and filtering out improper content in the received user voice input.

[0014] The improper content may comprise at least one of the following: inappropriate content, non-guidance instructions related content.

[0015] The determining the guidance action may comprise routing a path to a location of the determined user destination based on the building related data. The building related data may be pre-trained and / or retrieved from one or more building related databases and / or external information systems.

[0016] The building related data may comprise at least one of the following: floor plan(s) of the building, blueprint(s) of the building, floor map(s) of the building, location data of different spaces in the building, person data relating to the building, parking space related data, data relating to elevator systems in the building, calendar system data.

[0017] The at least one Al model may be at least one of: a generative Al model, a reasoning Al model, a tool / function / agent based Al model.

[0018] According to a second aspect, a processing system for providing user guidance in a building is provided, wherein the processing system is configured to perform the steps of the method described above.

[0019] According to a third aspect, a user guidance system for providing user guidance in a building is provided, wherein the user guidance system comprises: a user interface unit comprising: an input device for receiving a voice input from a user, and at least one output device for providing guidance to the user; and a processing system described above communicatively coupled to the user interface unit.

[0020] According to a fourth aspect, an elevator system is provided, wherein the elevator system comprises: at least one elevator car configured to travel along a respective at least one elevator shaft, an elevator control unit, and a guidance system described above.

[0021] According to a fifth aspect, a computer program product is provided, wherein the computer program product comprises instructions which, when the program is executed by a computer, cause the computer to carry out the method described above.

[0022] According to a sixth aspect, a computer-readable medium having stored thereon the computer program described above is provided.

[0023] Various exemplifying and non-limiting embodiments of the invention both as to constructions and to methods of operation, together with additional objects and advantages thereof, will be best understood from the following description of specific exemplifying and non-limiting embodiments when read in connection with the accompanying drawings.

[0024] The verbs “to comprise” and “to include” are used in this document as open limitations that neither exclude nor require the existence of unrecited features. The features recited in dependent claims are mutually freely combinable unless otherwise explicitly stated. Furthermore, it is to be understood that the use of “a” or “an”, i.e. a singular form, throughout this document does not exclude a plurality.

[0025] BRIEF DESCRIPTION OF FIGURES

[0026] The embodiments of the invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings.

[0027] Figure 1 illustrates schematically an example of a portion of an elevator system with a user guidance system.

[0028] Figure 2 illustrates schematically an example of a user interface unit of the user guidance system.

[0029] Figure 3 illustrates schematically an example of a method for providing user guidance in a building.

[0030] Figure 4 illustrates schematically an example of the user interface unit receiving user voice input from a user.

[0031] Figure 5 illustrates schematically an example of components of a processing system.

[0032] DESCRIPTION OF THE EXEMPLIFYING EMBODIMENTS

[0033] Figure 1 illustrates schematically an example of a portion of an elevator system 100 with a user guidance system 200. The user guidance system 200 may be arranged at least functionally in connection with the elevator system 100. Alternatively or in addition, the elevator system 100 may comprise the user guidance system 200. In other words, the user guidance system 200 may be part of the elevator system 100.

[0034] The elevator system 100 comprises at least one elevator car 110 configured to travel along a respective elevator shaft 120 between a plurality of landings (i.e. floors). The elevator system 100 further comprises an elevator control unit 130. The elevator control unit 130 may be configured to control one or more operations of the elevator system 100. The elevator control unit 130 is communicatively coupled to the other entities of the elevator system 100. The communication between the elevator control unit 110 and the other entities of the elevator system 100 may be based on one or more known communication technologies, either wired or wireless. The implementation of the elevator control unit 130 may be done as a standalone control entity or as a distributed control environment between a plurality of stand-alone control entities providing distributed control resource. The elevator system 100 may further comprise one or more known elevator related entities, e.g. a counterweight, an elevator hoisting machine, elevator doors, and / or safety circuit and devices, etc., which are not shown in Figure 1 for sake of clarity.

[0035] The user guidance system 200 comprises a user interface unit 210. The user interface unit 210 may be arranged at any floor of the building. Preferably, the user interface unit 210 is arranged at a lobby or entry floor of a multistorey building. Figure 2 illustrates schematically an example of the user interface unit 210 of the user guidance system 200. The user interface unit 210 comprises an input device 212 for receiving a voice input from a user 140 and at least one output device 214a, 214b for providing guidance to the user 140. The input device 212 of the user interface unit 210 may comprise an audio sensor, e.g. a microphone. The at least one output device 214a, 214b of the user interface unit 210 may comprise at least one speaker 214a, at least one display 214b, and / or at least one projector.

[0036] The user guidance system 200 comprises further a processing system 220 communicatively coupled to the user interface unit 210. The implementation of the processing system 220 of the user guidance system 200 may be done as a standalone processing entity or as a distributed processing environment between a plurality of stand-alone processing entities providing distributed processing resource. The processing system 220 of the user guidance system 200 is communicatively coupled to the elevator control unit 130 of the elevator system 100. The communication between the processing system 220 and the elevator control unit 130 of the elevator system 100 may be based on one or more known communication technologies, either wired or wireless. The processing system 220 may be implemented as a local processing system 220. For example, the user interface unit 210 may comprise the processing system 220 and / or the processing system 220 may be integrated in the user interface unit 210. Alternatively or in addition, the user interface unit 210 may be connected to an external server, such as a cloud server. The external server may comprise the processing system 220. The communication between the processing system 220 and the other components of the user interface unit 210 may be based on one or more known communication technologies, either wired or wireless. Thus, the processing system 220 may be local, external or a combination thereof. For example, one or more operations of the processing system 220 may be performed on the external server and / or one or more operations of the processing system 220 may be performed locally.

[0037] The user guidance system 200 may further comprise one or more databases 150. The one or more databases 150 may for example comprise one or more building related databases 152 and / or external information systems 154. Building related data may be stored into the one or more building related databases 152 and / or external information systems 154. Some non-limiting examples of the external information systems 153 may comprise one or more map services (either web- or map-based), such as Google maps.

[0038] Next an example of a method for providing user guidance in a building is described by referring to Figure 3, which illustrates schematically the method as a flow chart. The method is performed by the user guidance system 200 discussed above. The method may be a computer-implemented method performed by the the guidance system 200.

[0039] The user interface unit 210 receives by the input device 212 user voice input from the user 140. The user voice input is provided by the user 140 speaking out loud. The processing system 220 receives 310 the user voice input from the input device 212. The received user voice input may comprise an indication of a user destination, i.e. a destination the user 140 wants to reach. The indication of the user destination may be any kind of indication based on which the processing system 220 is capable of determining the respective user destination. The determination of the user destination based on the received user voice input is discussed more later in this specification. For example, the user 140 may say as the user voice input “I would like to go to a meeting room named Superstar” or “I have a meeting with company X” in an office building, “I am going to a movie theatre” or “I am going to store X”, In a shopping centre, or “I am going to see Mr. Smith” or “I am going to apartment X” in a residential building. The above-mentioned examples of the user voice input are only nonlimiting examples, and the user voice input may comprise any other indication of the user destination. The received user voice input may further comprise an indication of one or more conditions relating to a guidance action to be determined. For example, the received user voice input may comprise an indication how the user prefers to reach the user destination, e.g. the user 140 prefers to use elevator to reach the user destination or the user 140 prefers not to use elevator to reach the user destination, etc. The determination of the guidance action is discussed more later in this specification. Figure 4 illustrates schematically a non-limiting example of the user interface unit 210 receiving the user voice input from the user 140. In the example of Figure 4, the user 140 says for the user voice input “I am going to apartment A14”. There may, as shown in Figure 4, be instructions provided, such as shown on the display screen 214b and / or spoken out loud via the speaker 214a, by the user interface unit 210 even before the user 140 gives the user voice input.

[0040] The processing system 220 applies, e.g. executes, at least one artificial intelligence (Al) model at least for determining 320 a user destination from the user voice input and determining 330 a guidance action for guiding the user 140 to the user destination based on building related data. In other words, the processing system 220 feeds the received user voice input to the at least one Al model 526 as input to determine the user destination and the respective guidance action as outputs of the at least one Al model 526. As discussed above the user voice input comprises an indication of the user destination. The processing system 220 determines, e.g. identifies, the user destination indicated in the received user voice input by applying the at least one Al model 526. The at least one Al model 526 may be combined (e.g. integrated) with one or more known speed recognition technologies or virtual assistants (e.g. Amazon Polly, Amazon Alexa, Google Assistant, etc.) to understand and process the user voice input. The at least one Al may be capable of understanding different languages. In other words, the at least one Al is capable of determining the user destination and the respective guidance action regardless of the language used in the user voice input. This enables that the user guidance system 200 supports the use of multiple languages, i.e. the user guidance system 200 is multilingual. This is turn enables the user guidance system 200 meets the needs of an international user base. The determining the guidance action may for example comprise routing a path to a location of the determined user destination (i.e. a path between a location of the user interface unit 210 and the location of the determined user destination) based on the building related data. The routing of the path may comprise identifying the location of the determined user destination, e.g. address or coordinate information about the location of the determined user destination. The location of the determined user destination may be on the same floor as the location of user interface unit 210. In that case, the path to the location of the determined user destination does not comprise use of elevator cars, escalators, and / or stairs. Alternatively, the location of the determined user destination may be on another floor than the location of the user interface unit 210. Thus, the path to the location of the determined user destination may comprise use of elevator cars, escalators, and / or stairs. The user destination may for example be a space, a room, an apartment, or a parking space, etc. within the building.

[0041] For sake of completeness, it is worthwhile to mention that the applied at least one Al model 526 to generate the output may be any type of model suitable for the task. For example, a generative Al model may be used, but also a reasoning Al model and / or a tool / function / agent based Al model (aka Al agents) are applicable. Moreover, the at least one Al model 526 may be implemented by harnessing a plurality of Al models to cooperate in the generation of the output.

[0042] The building related data may for example pre-trained and / or retrieved from one or more building related databases 152 and / or external information systems 154. When the building related data is pre-trained, the building related data may be used in the training of the at least one Al model. In other words, the building related data may be used as training data of at least one pretrained Al model 526. The at least one pre-trained Al model may then be used as the at least one Al model 526. The use of the at least one pre-trained Al model 526 improves the performance and / or the reliability of the user guidance system 200. The use of the at least one pre-trained Al model 526 may also reduce the risk of hallucinations in the determined guidance action. When the building related data is retrieved from the one or more building related databases and / or external information systems, the retrieved building related data may be used in the determination of the guidance action as further input data for the at least one Al model 526. Thus, instead of or in addition to pre- training the at least one Al mode, the at least one Al model 526 may use the retrieved building related data as input data used for the determination of the guidance action. The building related data may be retrieved in real-time. To be able to retrieve the building related data, the at least one Al model 526 has access to the one or more building related databases 152 and / or external information systems 154 in real-time.

[0043] The building related data may for example comprise at least one of the following: floor plan(s) of the building, blueprint(s) of the building, floor map(s) of the building, location data of different spaces in the building (e.g. rooms, offices, apartments, etc.), person data relating to the building (e.g. data relating to residents, occupants, tenants, employees, companies, etc. of the building), parking space related data (e.g. location data of parking places, image data from a parking garage, etc.), data relating to elevator systems in the building, calendar system data (e.g. calendar systems of meeting rooms). As mentioned above, the guidance action for guiding the user 140 to the user destination is determined based on the building related data. The processing system 220 may for example apply the at least one Al model 526 to identify the location of the user destination and to determine the path to the location of the user destination based on the building related data, e.g. by utilizing the floor plan(s) of the building, location data of different spaces in the building, the data relating to the elevator systems in the building, and / or any other kind of building related data. The data relating to the elevator systems in the building may be required for the determining the path to the location of the user destination, when the location of the user destination is on a different floor than the user 140 is currently located.

[0044] According to an example, the user voice input may comprise an indication of a specific room or apartment in the building. In that case the processing system 220 may first determine from the user voice input by applying the at least one Al model 526 that the user destination is the specific room or apartment indicated in the received user voice input. Then the processing system 220 may identify (e.g. locate) by applying the at least one Al model 526 the location of the room or apartment based on the building related data comprising e.g. floor plan(s), blueprint(s) and / or floor map(s). When the location of the room or apartment is located, the processing system 220 may determine by applying the at least one Al model 526 the path to the location of said room or apartment based on the building related data comprising e.g. location data of different spaces in the building and / or data relating to elevator systems in the building.

[0045] According to another example, the user input may comprise an indication of the name of a person living in a residential building or working in an office building. In some examples, the user voice input may alternatively or in addition comprise an indication of the name of the company residing in the office building. In case the user voice input comprises the indication of the name of the person living in a residential building, the processing system 220 may first determine from the user voice input by applying the at least one Al model 526 that the user destination is an apartment in which the person indicated in the received user voice input lives. Then the processing system 220 may identify (e.g. locate) by applying the at least one Al model 526 the location of the apartment based on the building related data comprising e.g. floor plan(s), blueprint(s), floor map(s), and / or person data relating to the building. The person data relating to the building may for example comprise data indicating the apartment in which said person lives. When the location of the apartment is located, the processing system 220 may determine by applying the at least one Al model 526 the path to the location of said apartment based on the building related data comprising e.g. location data of different spaces in the building and / or data relating to elevator systems in the building.

[0046] According to yet another example, the user voice input may comprise an indication of a license number of a car of the user 140 parked in a parking garage of the building. In that case the processing system 220 may first determine from the user voice input by applying the at least one Al model 526 that the user destination indicated in the received user voice input is a parking space to which the car is parked. Then the processing system 220 may identify (e.g. locate) by applying the at least one Al model 526 the location of the parking space within the parking garage of the building based on the building related data comprising at least the parking space related data. When the parking space is identified, the processing system 220 may determine by applying the at least one Al model 526 the path to the location of said parking space based on the building related data comprising e.g. location data of different spaces in the building and / or data relating to elevator systems in the building. The processing system 220 may further apply the at least one Al model 526 for pre-processing the received user voice input for detecting and filtering out improper content in the received user voice input. The improper content may for example comprise at least one of the following: inappropriate content, nonguidance instructions related content. The pre-processing is illustrated with an optional step 350 in Figure 3. The inappropriate content may for example comprise swear words, offensive phrases, and / or any other kind of inappropriate content. The detecting and filtering out the inappropriate content in the received user voice input may for example comprise creating an extensive vocabulary and / or a collection of phrases that includes inappropriate words and phrases and filtering out these words and phrases from the received user voice input before using the user voice input in the determination of the user destination. The detecting and filtering out the inappropriate content in the received user voice input may alternatively or in addition comprise utilizing machine learning and / or natural language processing models to detect the inappropriate content in the received user voice input. The non-guidance instruction related content may comprise any content that is not relevant for the determination of the user destination. The detecting and filtering out the non-guidance instructions related content may for example comprise limiting the at least on Al model 526 to respond only to inputs that are related to navigation within the building. In other words, the at least one Al model 526 may be limited to determine the user destination and the respective guidance action from user voice input that is related to the navigation within the building. This may reduce the risk of hallucinations and / or defects in the determined guidance action.

[0047] Alternatively or in addition, the processing system 220 may further apply the at least one Al model 526 for post-processing the determined guidance action for detecting and filtering out (i.e. removing) potential hallucinations in the determined guidance action. Outputs produced by the at least one Al model 526 that are factually incorrect and / or illogical (e.g. nonsensical) may be referred to as hallucinations. The hallucinations may for example comprise inconsistencies, factual errors, formatting errors, terminological errors, and / or misleading content, etc. in the generated guidance action. Filtering out the hallucinations from the determined guidance action improves the correctness and the reliability of the determined guidance action. The post-processing of the determined guidance action may for example comprise at least one of the following post-processing operations: checking the determined guidance action against the building related data, performing a consistency check, performing a redundancy and repetition check, performing a contextual verification, performing a temporal verification, comparing guidance actions determined by multiple Al models 526, performing language and grammar correction.

[0048] Checking the determined guidance action against the building related data may for example comprise checking (e.g. by one or more algorithms) the determined guidance action against known facts, e.g. against the building related data. Performing the consistency check may for example comprise checking that the determined guidance action is internally consistent. Performing the redundancy and repetition check may for example comprise detecting and removing redundant or repetitive information in the determined guidance action. This helps in producing more concise and coherent guidance action. Performing the contextual verification may for example comprise ensuring that the determined guidance action makes sense (e.g. is relevant and / or appropriate) within the given context (i.e. the user guidance). The temporal verification may for example comprise checking the timeliness of the information to ensure that the determined guidance actions are up-to-date and not outdated. As mentioned above one example of the post-processing step for detecting and filtering out possible hallucinations in the determined guidance action is comparing the guidance actions determined by multiple Al models 526. If the comparison of the guidance actions determined by multiple Al models 526 indicates that the guidance actions are consistent with each other, the guidance actions are more likely to be accurate. While discrepancies in the guidance actions may indicate hallucinations in guidance actions determined by one or more of the Al models 526. Performing the language and grammar correction may for example comprise using one or more other Al models or tools to correct grammatical errors and improve the comprehensibility of the determined guidance action. This may also help in identifying and correcting illogical outputs. The post-processing is illustrated with an optional step 360 in Figure 3.

[0049] According to an example, in response to a detection of the hallucinations in the determined guidance action, the processing system 220 may further generate a command to the at least one output device 214a, 214b to provide a request for the user 140 for providing further voice input comprising further indication of the user destination. After receiving the further voice input from the user 140 via the input device 212 of the user interface unit 210, the processing system 220 may redetermine or improve the determined guidance action by utilizing the further voice input to remove the hallucinations detected in the previously determined guidance action. The previously determined guidance action may then be replaced by the redetermined or improved guidance action. The redetermined or improved guidance action may further be post-processed for detecting and filtering out potential hallucinations in the redetermined or improved guidance action before replacing the previously determined guidance action. The replacement of the previously determined guidance action with the redetermined or improved guidance action means that the redetermined or improved guidance action may be used instead of the previously determined guidance action in the subsequent steps of the method, e.g. steps 340 and / or 370.

[0050] The same at least one Al model 526 may be used in the different processing steps 380 (e.g. pre-processing 350, determining 320 the user destination, determining 330 the guidance action, and / or post-processing 360) performed by applying the at least one Al model 526. For example, the same at least one Al model may be used in the pre-processing 350, in the determination 320 of the user destination, in the determination 330 of the guidance action, and in the post-processing 360. Alternatively or in addition, different at least one Al model 526 may be used at least in some of the processing steps 380. For example, different at least one Al model 526 may be used in each processing step 380, i.e. in the determination 320 of the user destination, in the determination 330 of the guidance action, and in the post-processing 360. According to another example, the at least one Al model 526 used in the determination 320 of the user destination and in the determination 330 of the guidance action may be the same at least one Al model and the at least one Al model 526 used in the pre-processing 350 and in the post-processing 360 may differ from the at least one Al model used in the determination 320 of the user destination and in the determination 330 of the guidance action.

[0051] The processing system 220 generates 340 at least one guidance command to the at least one output device 214a, 214b of the user interface unit 210 to provide guidance for the user 140 in accordance with the determined guidance action. In response to receiving the at least one guidance command from the processing system 220, the at least one output device 214a, 214b of the user interface unit 210 may provide (e.g. output) guidance for the user 140 in accordance with the determined guidance action. The guidance may comprise visual guidance and / or audible guidance. The visual guidance may be shown by the display screen 214b and / or the at least one projector. The audible guidance may be provided by the speaker 214a. The visual guidance may for example comprise one or more pictures, icons, and / or animations. In case the at least one output device 214a, 214b comprises the at least one projector the visual guidance may be projected on the floor(s) of the building. For example, a kind of a map of the building together with instructions to the user destination may be shown by the display screen 214b. Alternatively or in addition, the visual guidance may comprise textual instructions. For example, there may be a text shown on the display screen 214b “use the elevator A to go to the firth floor and then walk 20 meters to the right”. The audible guidance may for example comprise audible instructions to the user destination, i.e. the instructions to the user destination outputted by spoken words. For example, there may be a voice saying “use the elevator A to go to the firth floor and then walk 20 meters to the right”. As discussed above, the path to the user destination may comprise use of elevator, escalators, and / or stairs. Thus, also the guidance to the user destination may comprise instruction to use elevator, escalators, and / or stairs to reach the user destination. If the user destination cannot be reached by the closest elevator, the guidance may comprise instruction to the correct elevator to be used to reach the user destination. The guidance to the user destination may also comprise instruction how to reach the user destination without using elevator, e.g. in cases where the user destination is inaccessible with the elevator (e.g. a destination that the elevator cannot reach) or in cases where the user the user voice input indicates that the user 140 prefers not to use elevator to reach the user destination. According to an example, if the user voice input comprises a license number of a car of the user 140 parked in a parking garage of the building. The guidance to the user destination may comprise instructions to the parking space, where the car is parked. An example of the determination of the guidance action, e.g. the path to the location of the parking space, is discussed above.

[0052] As discussed above, the user guidance system 200 may be multilingual. To enhance the multilingualism of the user guidance system 200, the at least one Al model 526 may be capable of determining the guidance action so that the guidance provided in accordance with the determined guidance action may be provided with different languages. The guidance may for example be provided with the same language as the user voice input is given. Alternatively, the user voice input may comprise an indication of the language to be used for the guidance.

[0053] The processing system 220 may further generate 370 at least one elevator command to the elevator control unit 130 of the elevator system 100 to control at least one operation of the elevator system 100 in accordance with the guidance action to assist the user 140 to reach the user destination. In response to receiving the at least one elevator command, the elevator control unit 130 of the elevator system 100 may control the at least one operation of the elevator system 100 in accordance with the guidance action to assist the user 140 to reach the user destination. The at least one operation of the elevator system 100 may for example comprise driving the elevator car 110 in order to transport the user 140 from the current floor to a floor in which the user destination locates.

[0054] According to an example, the processing system 220 may further receive user feedback e.g. via the input device 212 of the user interface unit 210, that may be analyzed and / or used for further training of the at least one Al model 526 to improve the ability of understanding the user voice input by the at least one Al model 526 and the accuracy and reliability of the generated guidance action, which in turn improves the reliability of the user guidance system 200. Alternatively or in addition, the processing system 220 may further record and analyze interactions between the user interface unit 210 and the user 140 (e.g. the received user voice inputs and / or the determined guidance actions) that may also be used to further training of the at least one Al model 526. This may also improve the ability by the at least one Al model 526 of understanding the user input and the accuracy and reliability of the generated guidance action, which in turn improves the reliability of the user guidance system 200.

[0055] Figure 5 illustrates schematically an example of components of the processing system 220. The processing system 220 may comprise a processor unit 510 comprising one or more processors, a memory unit 520 comprising one or more memories, and a communication unit 530 comprising one or more communication devices. The mentioned elements may be communicatively coupled to each other with e.g. a communication bus. The memory unit 520 may store and maintain portions of a computer program (code) 525, the Al model 526, and any data. The computer program 525 may comprise instructions which, when the computer program 525 is executed by the processor unit 510 of the processing system 220 may cause the processor unit 510, and thus the processing system 220 to carry out desired tasks, e.g. one or more of the method steps described above. The processor unit 510 may thus be arranged to access the memory unit 520 and retrieve and store any information therefrom and thereto. For sake of clarity, the processor herein refers to any unit suitable for processing information and control the operation of the processing system 220, among other tasks. The operations may also be implemented with a microcontroller solution with embedded software. Similarly, the memory unit 520 is not limited to a certain type of memory only, but any memory type suitable for storing the described pieces of information may be applied in the context of the present invention. The communication unit 530 provides one or more communication interfaces for communication with any other unit, e.g. elevator control system 130, the input device 212, the at least one output device 214a, 214b, one or more databases 150, 152, 154, and / or with any other unit. The computer program 525 may be a computer program product that may be comprised in a tangible nonvolatile (non-transitory) computer-readable medium bearing the computer program code 525 embodied therein for use with a computer, i.e. the processing system 220.

[0056] The method and the guidance system 200 discussed above enable that the users can reach the user destination effortlessly without having to remember the floor or other location information of the user destination. The method and the guidance system 200 discussed above reduces time of the users as it provides a fast and efficient way to guide users to reach their destination. This is especially beneficial in complex environments, such as shopping centers or office buildings. As the user input and the guidance are provided in a noncontact manner and the elevator commands may be generated based on the user input, the users of the elevator system do not need to use any elevator buttons. The non-contactless improves also the hygiene and thus also prevents spread of diseases. The use of the guidance system 200 discussed above is easy and enables also easy of the elevators especially for persons to whom the use of traditional elevators is challenging.

[0057] The specific examples provided in the description given above should not be construed as limiting the applicability and / or the interpretation of the appended claims. Lists and groups of examples provided in the description given above are not exhaustive unless otherwise explicitly stated.

Claims

CLAIMS1. A method for providing user guidance in a building, the method comprising: receiving (310) user voice input from an input device (212) of a user interface unit (210), applying at least one artificial intelligence (Al) model (526) for: determining (320) a user destination from the voice input; and determining (330) a guidance action for guiding the user (140) to the user destination based on building related data, and generating (340) at least one guidance command to at least one output device (214a, 214b) of the user interface unit (210) to provide guidance for the user (140) in accordance with the guidance action.

2. The method according to claim 1 , further comprising applying the at least one Al model (526) for post-processing (360) the determined guidance action for detecting and filtering out hallucinations in the determined guidance action, wherein the post-processing (360) of the determined guidance action comprises at least one of the following: checking the determined guidance action against the building related data, performing a consistency check, performing a redundancy and repetition check, performing a contextual verification, performing a temporal verification, comparing guidance actions determined by multiple Al models, performing language and grammar correction.

3. The method according to claim 2, further comprising in response to the detection of the hallucinations in the determined guidance action generating a command to the at least one output device to provide a request for the user (140) for providing further voice input comprising further indication of the user destination.

4. The method according to any of the preceding claims, further comprising generating (370) at least one elevator command to an elevator control unitelevator system (100) in accordance with the guidance action to assist the user (140) to reach the user destination.

5. The method according to any of the preceding claims, further comprising applying the at least one Al model (526) for pre-processing (350) the received user voice input for detecting and filtering out improper content in the received user voice input.

6. The method according to claim 5, wherein the improper content comprises at least one of the following: inappropriate content, non-guidance instructions related content.

7. The method according to any of the preceding claims, wherein the determining the guidance action comprises routing a path to a location of the determined user destination based on the building related data.

8. The method according to any of the preceding claims, wherein the building related data is pre-trained and / or retrieved from one or more building related databases (152) and / or external information systems (154).

9. The method according to any of the preceding claims, wherein the building related data comprises at least one of the following: floor plan(s) of the building, blueprint(s) of the building, floor map(s) of the building, location data of different spaces in the building, person data relating to the building, parking space related data, data relating to elevator systems in the building, calendar system data.

10. The method according to any of the preceding claims, wherein the at least one Al model (526) is at least one of: a generative Al model, a reasoning Al model, a tool / function / agent based Al model.

11. A processing system (220) for providing user guidance in a building, wherein the processing system (220) is configured to perform the steps of the method according to any of the preceding claims.

12. A user guidance system (200) for providing user guidance in a building, the user guidance system (200) comprises: a user interface unit (210) comprising:an input device (212) for receiving a voice input from a user (140), and at least one output device (214a, 214b) for providing guidance to the user (140); and a processing system (220) according to claim 11 communicatively coupled to the user interface unit (210).

13. An elevator system (100), comprising: at least one elevator car (110) configured to travel along a respective at least one elevator shaft (120), an elevator control unit (130), and a guidance system (200) according to claim 12.

14. A computer program product (525) comprising instructions to cause the processing system (220) according to claim 11 to execute the steps of the method according to any of claims 1 to 10.

15. A computer-readable medium having stored thereon the computer program of claim 14.

Images