A system for transporting objects using at least one reusable smart mobile container and a chatbot.

Abstract

This invention relates to an object transport system that enables interactive communication. According to the present invention, such a system comprises at least one intelligent mobile object transport container (1) configured to capture first data provided by an interlocutor, an identifier for an object present in the container, and an intelligent dialogue agent (18). The intelligent dialogue agent (18) is configured to receive the first data together with the object identifier, perform contextualized semantic analysis of the first data in light of the object identifier, optionally generate contextualized second data resulting from the semantic analysis, and optionally transmit the second data to the at least one container. The container (1) is further configured to process the received second data.

Description

Technical Field

[0001] The present invention relates to the field of interactive artificial intelligence related to connected objects known by the term IoT (Internet of Things).

[0002] More precisely, the present invention relates to intelligent containers intended for the transport of products, parts, or objects.

[0003] The present invention is particularly applied to the transport of packages or parcels, used in the field of logistics transport for the delivery of products.

Background Art

[0004] In the field of logistics, it is known to label parcels, identify their destinations, sort them, and transfer them to their recipients via a logistics platform. Such techniques are subject to the need to add numerous labels depending on the number of logistics service providers involved in the transport. Furthermore, this basic technology encompasses many drawbacks related to the lack of peeled labels, readability problems, and the impossibility of tracking parcels between two identification points on the logistics platform. As a result, many parcels are lost and, in some cases, stolen.

[0005] Therefore, there is a clear need to use packages equipped with embedded electronic units having screens that function as labels and can transmit location information to a monitoring platform along the logistics route to be followed.

[0006] Finally, in order to strengthen the security of these packages, parcel locking means have been added to allow opening under specific conditions and restrict access to authorized persons, very often the final recipient.

[0007] However, these developments in packages for sending and transporting objects do not have no drawbacks.

[0008] Specifically, as soon as a new type of labeling is used, it is necessary to revise and redesign the process and anticipate operational and processing errors in the logistics chain.

[0009] Another drawback of such technology is that, in turn, all professionals in the logistics chain are required to be trained in this new, and often specialized, use.

[0010] Another drawback of this technology is that when the recipient receives their package, they have no information about its transport, even if they were able to track its route. [Overview of the Initiative] [Problems that the invention aims to solve]

[0011] Accordingly, the present invention aims to mitigate the drawbacks of the prior art described above.

[0012] More precisely, the present invention aims to provide an object transport system that enables interactive communication between the user and the transport package.

[0013] Another objective of the present invention is to provide such a technology that is easy for users to implement.

[0014] Another object of the present invention is to provide such a technology that enables appropriate interaction with the user.

[0015] Another objective of the present invention is to provide a technology that enables dialogue centered on the sender's profile.

[0016] Another object of the present invention is to provide such technology in a reliable and consumer-safe manner. [Means for solving the problem]

[0017] These objectives, along with other objectives revealed in the remainder of this text, are achieved using an object transport system that enables interactive communication with the interlocutor.

[0018] In the context of the present invention, the system comprises at least one intelligent, reusable object transport parcel, also known as a container, which is suitable for containing an object and capable of interacting with an interlocutor. The parcel is configured to capture first data.

[0019] The system also includes object identifiers for the contained objects, in other words, for the objects present in the container during transport.

[0020] The system further, a) In a container, during transport, a first data paired with the object identifier contained within is received, the first data is transmitted by the container, and further, the object identifier is transmitted, enabling interactive communication regarding the transported object. b) Perform semantic analysis of the received first data, contextualized in light of the object identifier paired with the first data, and form semantic results. c) Optionally, generate a second contextualized data resulting from semantic analysis, d) optionally comprising an intelligent dialogue agent configured to transmit a second data to at least one container in response to a first data.

[0021] Furthermore, in this system, the container is configured to process the second received data.

[0022] Therefore, in such a system, these technical features enable the user to become the interlocutor of the package and initiate a dialogue. In this dialogue, the container sends data to an intelligent agent, which then semantically analyzes it and uses the results of that analysis to respond to the user.

[0023] According to a particularly advantageous embodiment of the present invention, the intelligent dialogue agent - selects a first corpus of initial conditions from a set of libraries or corpora of initial conditions according to selection criteria, - is configured to use the first corpus of initial conditions during the generation of the contextualized second data.

[0024] These technical features significantly reduce the amount of input data for the model and also provide the advantage of faster processing.

[0025] According to an advantageous embodiment of the present invention, the selection criteria depend on the object identifier and / or the result regarding the meaning of the analysis of the first data.

[0026] These technical features target the desired target data with a corpus of initial conditions selected according to the situation. This provides the advantage of enhancing the validity of the answers to the questions of the interlocutors using the intelligent package.

[0027] According to a preferred embodiment of the present invention, the first corpus of initial conditions includes information about the object to be transported, enabling the intelligent dialogue agent to hold a dialogue regarding the object.

[0028] These technical features enable the reusable intelligent object transport package to answer questions from the recipient about any topic related to the delivered product or object in a reasonable manner.

[0029] Furthermore, by selecting a specialized corpus, the reliability of the provided responses can be enhanced and the hallucination phenomenon can be restricted.

[0030] According to a particular embodiment of the present invention, the intelligent dialogue agent is configured to select from a library, according to selection criteria, at least one second corpus specialized for initial conditions, such that the first corpus and at least one second corpus form a group of corpora.

[0031] According to an advantageous embodiment of the present invention, the intelligent dialogue agent is - Using filters based on selection criteria, extract a subset of data from a corpus group. -The system is configured to use a subset of the aforementioned data in the generation of a contextualized second set of data.

[0032] These technical features allow intelligent dialogue agents to adapt to a variety of situations. In particular, intelligent dialogue agents can handle situations where several objects are present in a package being delivered. In such situations, an object identifier is sent to each object contained in the package. In this way, it is possible to select, for example, one initial condition corpus for each object. However, the selection of multiple corpora related to the simultaneous delivery of several objects in the same package should not be understood as an limitation.

[0033] According to a particular embodiment of the present invention, the container is - Store the object identifiers of the objects present in the container, - It is configured to attach a stored object identifier to the first data sent to the intelligent dialogue agent, paired with it.

[0034] It is understood that a container is configured to store several object identifiers and handle the simultaneous transport of several objects.

[0035] According to an advantageous embodiment of the present invention, - The container is configured to classify its interlocutor into a typical dialogue profile and send that typical dialogue profile to the intelligent dialogue agent. - The intelligent dialogue agent is configured to adapt semantic analysis according to a typical dialogue profile.

[0036] These technical features allow intelligent dialogue agents to focus their responses according to the classification of their interlocutor.

[0037] According to a particularly advantageous embodiment of the present invention, the intelligent conversational agent is configured to limit the transmission of information during the conversation to information concerning the safe transport of objects if a typical conversational profile does not have the status of a container recipient.

[0038] These technical features allow intelligent conversational agents to limit the dissemination of information about the object being delivered to the recipient, thus preventing participants in the delivery chain from knowing its contents.

[0039] According to one embodiment, the present invention also provides an object transport system comprising a fleet of containers.

[0040] These technical features allow the system to benefit from reports of information about the entirety of the containers forming the fleet. This can improve responses, for example, related to the use of reusable intelligent object transport parcels.

[0041] According to one embodiment, the mobile container is a reusable, intelligent object transport package or parcel.

[0042] According to a preferred embodiment of the present invention, the first data and the second data are so-called communication data that form an interactive exchange. The first data comes from the container user. The second data processing by the container involves transmitting the second data to the user. The intelligent dialogue agent further, - Maintain contextual information based on interactive exchanges, - The system is configured to use contextual information during the semantic analysis.

[0043] These features allow users to query the container in natural language and, in return, receive assistance and / or additional information provided in natural language. Furthermore, users can interact with the container through a series of ordered questions and answers.

[0044] In addition, intelligent conversational agents retain a history of questions and answers exchanged with users, allowing them to refine their analysis and improve the accuracy of their responses.

[0045] According to an advantageous embodiment of the present invention, the first data is so-called informational data of the type of identifier, or physical quantity, or operating setting.

[0046] These features allow the intelligent dialogue agent to receive information about the container's context. This information can be single, paired, or combined to provide a variable level of accuracy regarding the collected information. The first so-called informational data may include, for example, only the internal temperature and / or the external temperature, and the internal and / or external atmospheric pressure. This may further include information related to its motion, such as acceleration or position determined by satellite positioning. Thus, from the motion information, the intelligent dialogue agent can infer that the container has been subjected to shock, vibration, or fall. In the case of a fall, the height of the fall and / or acceleration may complete the information.

[0047] Advantageously, the identifier is either an application-specific identifier or a known type of identifier, such as a SIM® (Subscriber Identification Module) card or an NFC® (Near Field Communication) card identifier.

[0048] According to a particularly advantageous embodiment of the present invention, the second data is a command resulting from the semantic analysis of the first so-called information data and applied by the container. These features allow the second data to be optionally generated and transmitted to the container, and the container to use it, depending on the result of the semantic analysis. Thus, the intelligent dialogue agent can choose to output text or voice messages, or command various actions, in particular, to control light-emitting diodes (LEDs), sirens or loudspeakers, motors, display screens, locking / unlocking devices. Thus, the intelligent dialogue agent can, for example, command a flashing mode for an LED to attract the user's attention.

[0049] According to a preferred embodiment of the present invention, the intelligent dialogue agent is -The first so-called information data is stored in a continuous learning parameter database, - The system is configured to train a model of an intelligent dialogue agent using the continuous learning parameters stored in the database.

[0050] These features allow intelligent conversational agents to increase their knowledge base, improve the quality of their responses, and / or enhance their validity.

[0051] According to a particular aspect of the present invention, training is automated.

[0052] According to a particularly advantageous embodiment of the present invention, the training is deep learning, and each continuous learning parameter has an associated weighting of positive or negative approval of the parameter.

[0053] Therefore, by using machine learning or deep learning type algorithms, the value of the training data is evaluated, and the reliability of subsequent semantic analysis is improved.

[0054] According to one embodiment, the container includes means for inputting the first so-called communication data, which is voice capture means and / or text input means.

[0055] These features allow users to communicate with the container by directly querying it verbally in natural language and / or by using input means such as a touch-sensitive or mechanical keyboard or a stylus associated with handwriting recognition. According to certain embodiments of the present invention, the input means is an independent short-range remote control device such as a smartphone, which communicates using short-range wireless communication means such as Bluetooth®, BLE® (Bluetooth Low Energy), or Wi-Fi® protocols.

[0056] These features allow for the manufacture of containers at a low cost, thereby expanding the target customer base for the product.

[0057] According to an advantageous embodiment of the present invention, the system comprises a text transcription means for the speech capture so that an intelligent dialogue agent receives the first data in text format.

[0058] These features allow users to communicate with the container by speaking, and the voice messages it captures are received by an intelligent conversational agent and converted into text messages that can be directly interpreted by conversational artificial intelligence.

[0059] According to one embodiment, the container includes the second so-called communication data audio playback means and / or display means.

[0060] These features allow users to receive answers to queries they indirectly submit to an intelligent dialogue agent via a container.

[0061] According to an advantageous embodiment of the present invention, the system comprises a means for transcribing second so-called communication data, which is generated in text format by an intelligent dialogue agent. These features allow the user to receive a voice response.

[0062] According to one embodiment, the container is selected from a group that includes luggage and reusable boxes.

[0063] Therefore, in novel and non-obvious ways, reusable boxes, reusable logistics delivery packages, electronic logistics labels intended to be equipped on disposable packages, and luggage such as shipping cases will enable users to interact with various types of information, particularly information that facilitates the use of containers.

[0064] In a particularly advantageous embodiment of the present invention, the intelligent dialogue agent is integrated within the embedded computer unit of the container and / or on a central computer platform configured to communicate with the container.

[0065] These features make it possible to have either a standalone container with localized intelligence or a container supported by intelligence transferred to a remote central computing platform. Furthermore, it is possible to distribute or concentrate the intelligence on either the central computing platform or the container to improve the validity and / or speed of responses, or to distribute the processing load according to the amount of processing power incorporated into the container.

[0066] In addition, the localized intelligence within each container also offers the advantage of improving confidentiality and / or privacy, especially in the case of hand luggage.

[0067] Furthermore, such localized intelligence mitigates the lack of connectivity to the central computer platform in the event of failures in long-range communication means. This also makes it possible to reduce the power consumption of long-range communication modules. This, in turn, makes it possible to manage and therefore minimize 4G data consumption.

[0068] According to a particular embodiment of the present invention, the intelligent dialogue agent is moved to an external computer platform that communicates with the central computer platform.

[0069] Furthermore, the intelligent dialogue agents and containers are configured to communicate with each other via a central computer platform.

[0070] These characteristics allow intelligent dialogue agents to be moved to external platforms, such as those of specialized providers.

[0071] According to a particular embodiment of the present invention, the transport system further comprises means for communicating with a global computer network. According to a modified embodiment, the communication means connects a central computer platform and / or external computer platforms to the global computer network.

[0072] These features allow intelligent conversational agents to complete their learning by searching the internet. Furthermore, they can improve the quality of their responses with information collected from global computer networks, such as weather information related to temperature peaks recorded by containers at specific locations.

[0073] According to an advantageous embodiment of the present invention, - The container is configured to identify users, classify identified users into typical conversation profiles, and send typical conversation profiles to an intelligent conversation agent. - The intelligent dialogue agent is configured to adapt semantic analysis according to a typical dialogue profile.

[0074] These features allow the intelligent conversational agent to provide targeted information according to typical conversational profiles, and therefore, according to the user's role, such as a customer receiving a package, a logistics worker responsible for transporting it, or an agent responsible for maintaining a reusable package. Thus, the identified maintenance agent can use voice command keywords to obtain diagnostic information or information or recommendations from after-sales service manuals that are not available to other roles.

[0075] According to one embodiment of the present invention, user identification is achieved through authentication.

[0076] Therefore, this technology can be used to ensure, for example, that only the recipient customer can open the container during delivery.

[0077] According to a particularly advantageous embodiment of the present invention, the intelligent dialogue agent is configured to select a dialogue language in accordance with the location of a container received in first so-called information data. The location, which is a physical positioning quantity, is typically the most recent GPS location of the container. However, it can also be estimated from other information, such as the current cell phone cell, or the known location of the current storage warehouse and / or its Wi-Fi® network to which the container can connect.

[0078] Therefore, the intelligent dialogue agent will converse in the local language if available, and otherwise switch to an international language, typically English, Spanish, or French.

[0079] Advantageously, intelligent dialogue agents use measured location to determine local cultural standards and adapt the generated responses to these standards.

[0080] Certain aspects of the present invention are based on the idea of ​​leveraging advances in artificial intelligence in the dialogue field, particularly using "generative pre-trained transformers" type LLMs (Large-Scale Language Models), such as OpenAI®'s GPT3® and ChatGPT®, Google®'s Bert® and Bard®, or using generative artificial intelligence.

[0081] These aspects of the present invention encompass the idea that responses provided to an interlocutor can be quickly and easily adapted or updated, particularly depending on the object being transported. Other aspects of the present invention include the idea of ​​significantly optimizing the validity of these responses to provide a high level of accuracy.

[0082] One objective of the present invention is to provide a technique that is reliable and, consequently, reduces the risk of hallucination with respect to the response provided. For the sake of clarity, in artificial intelligence, hallucination is the fact that a statistically plausible but false response is provided to the requester, who may perceive it as a seemingly real or certain fact.

[0083] Advantageously, the present invention utilizes RAG (Search Expansion Generator). Such generative artificial intelligence optimization is carried out using fundamental elements of technologies well known to those skilled in the art, such as Microsoft® AutoGen®, Langchain Inc.® Langchain®, and LlamaIndex®, which are cited as non-limiting examples useful for explanation.

[0084] Therefore, the use of such techniques has many advantages. Its use improves the effectiveness of LLM use in processing significantly or much smaller databases. This use also provides a high degree of adaptability to the creation of new corpora or the addition of new documents. Furthermore, it enhances the reliability of results through better control of the knowledge base by controlling the document sources used, thereby reducing the risk of hallucination phenomena regarding the responses provided. Such use further contributes to increased confidence in the responses provided due to the aforementioned advantages.

[0085] Other features and advantages of the present invention will become clearer upon understanding the following description of embodiments of the invention given as simple, explanatory, and non-limiting examples, as well as the accompanying drawings. [Brief explanation of the drawing]

[0086] [Figure 1] Figure 1 is a schematic diagram of an exemplary embodiment of an object transport system featuring a communication container according to the present invention. [Figure 2] Figure 2 is a schematic diagram of a second exemplary embodiment of intelligent baggage according to the present invention. [Modes for carrying out the invention]

[0087] 1. Exemplary Embodiments of the Invention Figure 1 shows a first embodiment of an object transport system according to the present invention, comprising a container 1, a computer infrastructure 14, and an external computer platform 19.

[0088] The interactive container 1 includes an embedded computer unit 2, data storage means 3, power supply battery 4, microphone 5, loudspeaker 6, digital screen 7, one or more control buttons 8, first communication means 9, and one or more actuators (not shown), and the whole is assembled using essentially known technology.

[0089] It further includes numerous sensors (not shown) to obtain various information related to its storage, transport, or environment. Thus, it may obtain GPS (Global Positioning System) location and motion information used to track its movement, estimate when it has been subjected to impact, estimate whether it has been exposed to vibration, or record drops. Container 1 may also record internal and / or external information such as temperature and atmospheric pressure. These various pieces of information may also include an identifier unique to Container 1, an identifier unique to Computer Unit 2, a SIM (Subscriber Identification Module) card identifier, an NFC (Near Field Communication) code, or any other known identifier essentially known to those skilled in the art.

[0090] The computer infrastructure 14 includes a computer platform 15, a database 16, and a second communication means 17 that communicates with the computer unit 2 of the container 1 via a first communication means 9.

[0091] Furthermore, the computer infrastructure 14 can communicate with the external computer platform 19 using the third communication means 20.

[0092] In this embodiment, the external computer platform 19 is a) Receive the captured data transmitted by container 1 via the computer infrastructure 14, b) Perform semantic analysis on the captured data, c) Generate data resulting from semantic analysis, d) Conversational artificial intelligence modules, typically including software, comprising a conversational agent 18 configured to transmit the resulting data to container 1 via the computer infrastructure 14.

[0093] If the captured data is the first so-called communication data and the resulting data is the second so-called communication data, a dialogue begins. In this case, the first and second communication data are stored by the dialogue agent 18 using essentially known methods to preserve contextual information. As the dialogue progresses, the quality of the dialogue history, composed of this contextual information, improves.

[0094] To initiate a dialogue, the user presses button 8 on container 1. The computer unit then begins capturing speech using microphone 5 and records the voice message spoken by the user. It then sends this captured voice message, called the first captured voice message, to the computer infrastructure 14. The computer infrastructure 14 is equipped with means for converting speech to text, known as "speech-to-text" and "text-to-speech," and is responsible for transcribing the first voice message into a corresponding first text message. The computer infrastructure 14 then sends this first voice message to the dialogue agent 18 on the external computer platform 19.

[0095] In the dialogue, the resulting data is a second text message, which is then received by the computer infrastructure 14. This infrastructure converts the second text message into a second voice message and sends it to container 1. Container 1 then plays the second voice message to the user via the loudspeaker 6, allowing the user to continue the dialogue. Thus, through this series of interactions, container 1 engages in dialogue with the user.

[0096] Furthermore, since it is equipped with other information sources specified above, container 1 is also configured to transmit at least one of various pieces of information that can be used as captured data messages. For example, the transmitted, captured data may convey information about its GPS location. Alternatively, the captured data may include a combination of information such as an identifier, internal temperature, and external temperature measured by container 1.

[0097] Therefore, if the captured data is so-called informational data rather than a user's voice message, the resulting data is a command sent to container 1, which then applies it to at least one actuator. For example, the resulting command might command the unlocking of container 1 and allow the user to open it. However, a user's voice message could be interpreted by the conversational agent 18 as a command to control one of the actuators. Nevertheless, in the configuration of container 1 and / or conversational agent 18, the user is not necessarily able to act on all the actuators equipped in container 1.

[0098] In the context of an object transport system, the set of data that is fed to the dialogue agent 18 by the fleet of containers contributes to improving the quality of the model.

[0099] Furthermore, contextualizing container 1 makes it possible to guide the conversation to provide the user with the most appropriate information possible regarding container 1, especially its potential.

[0100] Furthermore, contextual information is provided to the model of the dialogue agent 18 for training, giving it an initial corpus of conditions. Typically, this contextual information relates to container 1, its description, how to use it, possible uses, frequently asked questions from users, sales offers for transporting objects using container 1, or information about businesses that manage a fleet of containers 1 forming an object transport system.

[0101] To do this, the dialogue agent 18 undergoes a specialization of its model, in which it is provided with a set of keywords and associated semantics that form a corpus of initial conditions.

[0102] For example, learning about transportation-related events can be done by following the pseudocode provided below, which will be helpful in the explanation. "Each stored event is composed of type, value where date is the date of the event, type is the type of the event and value is defined per type as follows: - 'sender': name, address - 'recipient': name, address - 'startShipment': date, latitude,longitude - 'endShipment': date, latitude, longitude - 'sensors':date, 'externalTemperature',measurement (in °C), 'internalTemperature', measure (in °C),'externalHumidity', measure (in %), 'internalHumidity', measure (in %),'pressure', measure (in hPa) - 'geolocation': date, latitude, longitude - 'warehouse': enterDate, exitDate, latitude, longitude, name - 'flight': fromDate, fromLatitude,fromLongitude, fromAirport, toDate, toLatitude,toLongitude, toAirport - 'alert': date, latitude, longitude, subtype (in 'externalTemperature', 'internalTemperature', 'externalHumidity','internalHumidity', 'freeFall'), measure -'events': You will return every event containing the keyword 'store' as a list without displaying the keyword itself. When 'event' is followed by a type, return every events of the defined type as a list instead of all 'store' keywords and return the closest city instead of raw location coordinates." Learning related semantics can be done as follows: "A duration between two events event A and event B is defined as the difference between the date of the event B and the date of the event A. The duration of an event is defined as the difference between the exitDate and the enterDate of the event or the toDate and the fromDate of the event.” Container 1 is, for example, a transport parcel or package according to the present invention.

[0103] The principle of contextualization, centered on the use of containers, was explained earlier. In this example, the initial conditions corpus includes all or at least some information about the containers that is considered useful in answering user inquiries on this topic. In this case, the contextual information includes this type of initial conditions corpus specialized for information related to the containers, and the history of questions / answers exchanged with the interlocutor or user. Furthermore, the initial conditions corpus concerning the transport of containers is also explained, taking into account the information raised by container 1 and / or the fleet of container 1.

[0104] However, this principle of contextualization should not be understood as a restrictive claim, but merely as an example to aid in explanation. Therefore, other types of corpora are included within the scope of this invention.

[0105] In a particularly advantageous embodiment, a container 1 in the form of a reusable intelligent transport package has information about the object or product it contains and transports. Thus, contextual information may relate to any information specific to the object or product being transported, which contributes to the formation of a corpus of specific initial conditions used by the model of the conversational agent 18.

[0106] As a non-limiting example useful for explanation, this information concerning the object being transported may include detailed characteristics of the object, whether such characteristics are technical or aesthetic, advantages, optionally disadvantages or limitations, and even details of its implementation or use.

[0107] In other words, the computer infrastructure 14 and / or container 1 stores the object identifier of each object present in the container 1's containment space during transport. By communicating each object identifier to the model of the conversational agent 18, the conversational agent 18 determines the corpus of contextual data to be used from a library also known as a set of initial condition corpora. For example in this embodiment, it is understood that the initial condition corpus is specialized in particular for one object, and in this case the initial condition corpus is specialized. In other words, the conversational agent 18 selects an initial condition corpus relevant to the object being transported from a library containing a set of initial condition corpora, a set of objects with no essential connections whatsoever, using the object identifier, which serves as a selection criterion.

[0108] It should be understood that the initial conditions corpus relates to a broader range of things than a single object; for example, it may include descriptions of various products or a product catalog containing all the products that a seller can sell. It is also clear that information that can help improve the quality of the corpus can come from lists of recommended products or services directly or indirectly related to the products being transported, such as user instructions, repair conditions, or recommended maintenance procedures for products already mentioned.

[0109] Therefore, thanks to information about the product being transported, on the one hand through object identifiers and on the other through contextual information specific to the product being transported collected in the model of the dialogue agent 18, container 1 can engage in dialogue about the object it is transporting. It can inform the recipient about the product they will receive, using a knowledge base far exceeding that of a human dialogue partner acting as, for example, a sales agent, a technical dialogue partner, or an after-sales service representative. If the product being delivered is a gourmet product such as a Grand Cru wine, the receiving customer can, during the interactive exchange upon receipt, obtain advice and information about the wine received, such as its sensory characteristics or any other characteristics, food pairings or aging period, and conditions for storing and consuming this Grand Cru wine. Thus, this information or interactive exchange could be recommendations to the user, selling points or benefits, or words of congratulations for making the right choice in purchasing the product.

[0110] To do so, for example, container 1 receives identification information of the products it contains from a remote server, typically a computer infrastructure 14. Alternatively, using its built-in image sensor directed into the internal space of the container that can contain products or objects, the container captures at least one image to identify the products or objects. This identification is achieved by any means known to those skilled in the art, in particular by image recognition techniques embodied by, for example, the computer infrastructure 14. Any other known methods can be applied to enable the container to obtain an identifier of the object it is transporting, in particular via a label associated with the object that emits the identifier, in near-field communication techniques.

[0111] Therefore, container 1 and / or the computer infrastructure 14 can associate contextual information corresponding to the products present in container 1. This identifier is paired by container 1 and / or the computer infrastructure 14 with a first so-called communication data received by the conversational agent 18. In summary, a question made by the interlocutor in container 1 is sent to the conversational agent 18 in the form of first data associated with an object identifier, and the conversational agent 18 provides an answer via second so-called communication data. This answer provides information about the object being transported, the product, if the question relates to this topic.

[0112] In the context of this invention, it is perfectly clear and understood that at any point during travel, i.e., during the transport of container 1, information raised by the container and collected by the model of the conversational agent 18 can be transmitted, in whole or in part, to participants in the logistics chain, shippers, or recipients.

[0113] However, if confidentiality is required, the information remains shared between the current user and the conversational agent 18. Furthermore, traces of this contextual information are not retained after delivery. In other words, the object identifier in this particular situation is erased from one or more locations where it is stored. The relationship between this object identifier and the customer during this delivery does not persist and therefore no longer exists.

[0114] Furthermore, only information regarding property and personal safety is available for communication with various parties in the logistics chain during transit. Aside from this safety information, all other information is available solely to provide communication with the clearly identified recipient of Container 1, a reusable intelligent parcel. In other words, before transmitting information disclosing the nature of the object being transported, the type of interlocutor initiating the communication is identified and then categorized or classified into a typical communication profile to ensure that the interlocutor is indeed the recipient of Container 1. In other words, only the recipient of the parcel obtains information about the object they will receive during the communication. Therefore, once the parcel is dispatched, the recipient is the only person who can obtain all the information available in the corpus of initial conditions corresponding to the product being delivered.

[0115] In a modification of the previous embodiment, the dialogue agent 18 selects several initial condition corpora from the library to form a group of selected corpora. This selection is again carried out according to a selection criterion, which is the received object identifier. This selection criterion may also be according to the question asked, or according to a combination of the question and the object identifier paired with it. If the question asked is considered in the selection criterion, it is clear that the parameters of the selection criterion are the results of the semantic analysis of this question. Furthermore, after the selection of several initial condition corpora, a step follows in which a portion of the data is selected from the group of corpora to form a subset of the extracted data. In other words, this step corresponds to filtering the data present in the group of corpora using a filter based on the previously defined selection criterion. In this modification, the intelligent dialogue agent ultimately uses this subset of extracted data, which is associated with the question / answer history exchanged with the user, to generate its response.

[0116] As a result, the use of a library or set of corpora allows for the adaptation of implementation plans and, therefore, the selection of whether to share or separate data from different corpora, depending on the confidentiality requirements of the products being transported.

[0117] 2. Other exemplary embodiments of the present invention Figure 2 shows a container 1, which appears here in the form of a transport case 10, forming an object transport system according to a second embodiment of the present invention.

[0118] Here too, container 1 is equipped with electronic means including an embedded computer unit 2, data storage means 3, power supply battery 4, microphone 5, loudspeaker 6, and button 8, all assembled in accordance with essentially known technology. The embedded computer unit 2 of transport case 10 is configured to capture requests explicitly made by the user of transport case 10 using the microphone 5 and to emit a response via the loudspeaker 6. Thus, the microphone 5 and loudspeaker 6 form an interface with the user of transport case 10.

[0119] Furthermore, the embedded computer unit 2 is configured to run an intelligent dialogue agent and thus autonomously interact with the user of the transport case 10.

[0120] In addition, the embedded computer unit 2 can receive a specialized corpus, selected according to the intended use of the transport case 10, using the communication means 9. For example, this corpus is specialized for a particular trip by including information on planned transit points such as train stations, airports, and urban transportation infrastructure, as well as information on subway station maps. Thus, the user can obtain information to find their position in the labyrinth of transit points through interaction with the transport case 10.

[0121] In a modified version of the second embodiment, the embedded computer unit 2 communicates with an intelligent dialogue agent 18, which has been moved to a computer infrastructure 14 (not shown) or an external computer platform 19, using communication means 9.

[0122] In the context of this invention, it is of course intended that the principles described in the previous chapter concerning the interaction of objects contained in the intelligent transport parcel-type container 1 be applied to the baggage shown in Figure 2.

[0123] Therefore, based on the same operation, the baggage owner takes on the role of the parcel recipient and becomes the interlocutor of the intelligent dialogue agent 18.

[0124] In this case, the owner may notify the baggage of the object or product contained within the baggage if it cannot be obtained in accordance with the previously described method. Thus, the object identifier is known, available, and can be paired with the first communication data as already described.

[0125] When the baggage is used for a proposal by an owner who is a salesperson of a product such as a new sample of a medical prosthesis or a new drug, the intelligent dialogue agent 18 can engage in a dialogue with a medical professional, doctor, or surgeon by selecting an appropriate initial condition corpus and convey all relevant information about the proposed product to that person. Therefore, there are no longer any unresolved questions that the salesperson may not have been able to answer immediately. This time saving increases the appropriateness of the sales activity and avoids the need to make further requests to busy professionals to convey additional information that the salesperson may not have been able to provide at the time.

[0126] 3. Any other features and advantages of the present invention Furthermore, the following can be provided in the modified embodiments of the present invention described in detail above. - If the captured data is at least one of various pieces of information, the conversational agent 18 records the captured data in a database for storing continuous learning parameters and trains its model using a technique known as "machine learning". Thus, the conversational agent 18 can train its model on the one hand, and it is configured to train it automatically on the other hand. - The captured data is associated with positive or negative approval weights considered by the model of the conversational agent 18. Thus, the model training takes the form of "deep learning." This improves the statistical analysis performance of the model and, consequently, the validity of the conversational interaction with the user. - Integrate means of identifying and / or authenticating the user into the container to determine the user's role and, consequently, the dialogue profile to be adopted. Identification can be achieved by any known method, such as badges or speech recognition. - Integrate the speech-to-text conversion means into container 1, or transfer this means to an external platform or specialized provider. - The conversational agent 18 is integrated into the computer infrastructure 14, and response time is optimized in particular. - An interactive agent 18 is incorporated into the computer unit 2 of container 1, which functions as a transport package, and thus given autonomy to interact with the user. - An interactive agent 18 integrated into the computer unit 2 receives arbitrarily weighted continuous learning parameters from the computer infrastructure 14. The learning parameters relate to information feedback from the fleet of containers 1 currently in use. - Specialize the intelligent dialogue agent so that it adapts its semantic analysis to the current location of the container. - For example, in the event of a fall detection, a message requesting extra care in handling container 1 is sent, following the analysis of the captured data, based on a set value, in other words, a voice message for the user. - Use a touch-sensitive digital screen as the input and / or output interface between container 1 and the user. - The first communication means 9 is integrated into the electronic means of the transport case 10. - Depending on the customer's specifications, contextualization is performed, and the initial corpus used by the conversational agent 18 is divided to create a fleet specific to each customer of the logistics carrier. - As an identifier for the object, for example, use the serial number of the object being transported. - To facilitate the transport of parcel 1, we provide a specialized corpus for specific journeys that contains information about planned transit points such as train stations, airports, and urban transportation infrastructure. -An object identifier is considered a second identifier within the scope of the present invention if all other identifiers mentioned above are considered first identifiers, such as SIM® (Subscriber Identification Module) cards or NFC® (Near Field Communication) cards. -All identifiers previously defined and / or mentioned, in particular application-specific identifiers, are available for use by the intelligent dialogue agent. - The library or initial set of corpora is not necessarily fixed and can be dynamically constructed and / or qualitatively improved as needed. - Throughout the logistics chain, from sender to recipient, each user of a package is a dialoguer who wants to interact with an intelligent dialogue agent.

[0127] The aforementioned technology for manufacturing an object transport system that enables interactive interaction can be used with various types of containers, for example, to form cargo transport containers.

[0128] It should be understood that the elements shown in the diagram can be implemented in various forms of hardware, software, or combinations thereof. Preferably, these elements are implemented as a combination of hardware and software of one or more multipurpose devices programmed in an appropriate manner, which may include a processor, memory, and input / output interfaces.

[0129] This explanation is illustrative of the principles of this disclosure. Therefore, those skilled in the art will understand that, although not expressly described or shown herein, they can conceive of various designs that embody and fall within the scope of these principles.

[0130] Accordingly, it should be understood that, according to the principles of the present invention, one or more libraries or sets of initial condition corpora make it possible to select at least one corpus of contextual data that will serve as a medium for the conversational agent to interact with the user. These libraries may be qualitatively enhanced and / or specialized according to the requirements of other parties, such as the administrator of the object transport system or the shipper, and thus expand the context of possible discussions beyond the examples described above.

[0131] All examples and conditional terms used herein are intended for educational purposes to help readers understand the concepts provided by the inventors to advance the principles and techniques of this disclosure, and should be construed as not being limited to these specifically mentioned examples and conditions.

[0132] Furthermore, all statements relating to the principles, aspects, and embodiments of this disclosure, as well as specific examples thereof, are intended to encompass their structural and functional equivalents. Moreover, these equivalents are intended to include both currently known equivalents and future-developed equivalents, i.e., any developed elements that perform the same function regardless of their structure.

[0133] Accordingly, for example, a person skilled in the art will understand that the principle diagrams described herein represent conceptual diagrams of exemplary circuits that implement the principles of the present disclosure. Similarly, all block diagrams, flowcharts, and other diagrams may be represented in essence on a computer-readable medium and will therefore be understood to represent various processes that can be performed by a computer or processor, whether or not the computer or processor is explicitly depicted.

[0134] To the extent that embodiments of the present invention are described as being implemented at least in part by a software-controlled data processing device, it will be understood that machine-readable non-transient media for holding such software programs, such as optical discs, magnetic discs, semiconductor memory, or similar, are also considered to correspond to embodiments of the present invention.

[0135] The functions of the various elements shown in the diagram can be achieved through the use of dedicated hardware and hardware capable of executing software programs in conjunction with appropriate software programs. Where they are performed by processors, the functions can be performed by a single dedicated processor, a single shared processor, or multiple individual processors, some of which may be shared. Furthermore, the explicit use of the terms “processor” or “controller” should not be interpreted as referring only to hardware capable of executing software programs, but implicitly, without limitation, may include digital signal processors (DSPs), read-only memory (ROM), random access memory (RAM), and non-volatile storage media for storing software programs. Additionally, certain elements may be independent or combined into, for example, a microcontroller (MCU) or a system-on-a-component (SoC, an acronym for System-on-Chip).

[0136] Other conventional and / or personalized software programs may also be included. Similarly, all switches shown in the diagram are purely conceptual. Their functions may be performed by the operation of program logic, by dedicated logic circuits, by the interaction of program instructions and dedicated software programs, or manually, and specific techniques are selectable by the executor, so that they are understood more concretely in context.

[0137] In the claims of this document, any element represented as a means for performing a specified function is intended to encompass any method for performing the function, including, for example, a) a combination of circuit elements for performing the function, or b) any form of software program. The software program consequently includes a microprogram, microcode, or the like, and is combined with appropriate circuitry for executing the software program in order to perform the function. The disclosure as defined by such claims lies in the fact that functions provided by different referred means are combined and brought together in the manner defined in the claims. Accordingly, all means that may provide these functions are considered equivalent to those described herein.

Claims

1. An object transport system that enables interactive communication with an interlocutor, A system comprising at least one intelligent mobile object transport container configured to accommodate an object, interact with an interlocutor, and capture first data, During the transport of the object, the object identifier of the object contained in the container (1) and An intelligent dialogue agent (18) is provided, The intelligent dialogue agent (18) said above, a) In the container, during transport, the container receives the first data paired with the object identifier contained therein, the first data is transmitted by the container, and the object identifier is transmitted to enable interactive communication regarding the object being transported. b) Perform semantic analysis of the received first data, contextualized in light of the object identifier paired with the first data, and form semantic results. c) Optionally, generate a second contextualized data resulting from the semantic analysis, d) Optionally configured to transmit the second data to at least one container in response to the first data, The container is further configured as an object transport system to process the received second data.

2. The intelligent dialogue agent (18) said above, From the library or set of initial condition corpora, select the first initial condition corpus according to the selection criteria. The object transport system according to claim 1, configured to use the first corpus of initial conditions during the generation of the contextualized second data.

3. The object transport system according to claim 2, wherein the selection criteria depend on the results relating to the meaning of the object identifier and / or the analysis of the first data.

4. The object transport system according to claim 2 or 3, wherein the first corpus of the initial conditions includes information relating to the object to be transported, enabling the intelligent dialogue agent to maintain a dialogue relating to the object.

5. The object transport system according to any one of claims 1 to 4, wherein the intelligent dialogue agent (18) is configured to select from a library, according to selection criteria, at least one second corpus specialized for initial conditions, and the first corpus and the at least one second corpus form a group of corpora.

6. The aforementioned intelligent dialogue agent, Using a filter based on the aforementioned selection criteria, a subset of data is extracted from the corpus group. The object transport system according to claim 5, configured to use a subset of the data in generating the contextualized second data.

7. The aforementioned container (1) is The object identifier of the object present in the container is stored, The object transport system according to any one of claims 1 to 6, configured to attach a stored object identifier to the first data transmitted to the intelligent dialogue agent (18), paired with the first data.

8. An object transport system according to any one of claims 1 to 7, comprising a fleet of containers.

9. The container is configured to classify its interlocutor into a typical dialogue profile and to transmit the typical dialogue profile to the intelligent dialogue agent. The object transport system according to any one of claims 1 to 8, wherein the intelligent dialogue agent is configured to adapt the semantic analysis according to the typical dialogue profile.

10. The object transport system according to claim 9, wherein the intelligent dialogue agent (18) is configured to limit the transmission of information during the dialogue to information relating to the safe transport of the object if the typical dialogue profile does not have the status of the recipient of the container (1).

11. The first data and the second data are so-called communication data that form an interactive exchange. The first data is provided by the interlocutor of the container, and the processing of the second data by the container is to transmit the second data to the interlocutor. The aforementioned intelligent dialogue agent further, The context information based on the aforementioned interactive exchange is retained, An object transport system according to any one of claims 1 to 10, configured to use the context information during the semantic analysis.

12. The object transport system according to any one of claims 1 to 11, wherein the intelligent dialogue agent is configured to select a dialogue language according to the geographical location of the container.

13. The object transport system according to any one of claims 1 to 12, wherein the intelligent dialogue agent is integrated within the embedded computer unit (2) of the container and / or on a central computer platform (14) configured to communicate with the container.

14. The object transport system according to claim 13, wherein the intelligent dialogue agent is moved to an external computer platform (19) configured to communicate with the central computer platform, and the intelligent dialogue agent (18) and the container (1) are configured to communicate with each other via the central computer platform.

15. The object transport system according to any one of claims 1 to 14, wherein the movable container is a reusable intelligent object transport parcel (1).

Images