Computer-implemented method for determining an optimized item arrangement in a stowing space

Abstract

A computer-implemented method for determining an optimized item arrangement 10 in a stowing space 12. Preferably, the items are luggage and the stowing space is a vehicle trunk / boot 14. The method inv

Description

Computer-implemented method for determining an optimized item arrangement in a stowing space. TECHNICAL FIELD The invention relates to a computer-implemented method for determining an optimized item arrangement in a stowing space. The invention further relates to a data processing device, a system, a computer program, and a computer-readable data carrier. BACKGROUND The idea of hopping in a car and setting off on a road trip may be the ultimate vacation. However, the reality of a road trip is that it requires a significant amount of time in planning and arranging, especially for long trips that span several days and cover hundreds of kilometers. Packing and arranging luggage into the vehicle is a skill, not every individual is good at, and in no way should it hamper the joy of a road trip. Inefficiency at optimally arranging luggage can result in users being forced to leave behind some important items. Depending on the vehicle, the rear seats can be configured to make additional room for storing luggage. Although luggage carried for road trips is usually the same, every trip is also unique at the same time. Primary factors that influence the planning and arrangement of luggage for a road trip may be the following: user preferences and priorities, number of passengers, duration of travel, luggage dimensions, boot space, and rear seat configurations. Since vehicles are getting more and more personalized, it would be desirable to have the boot space of the vehicle also personalized according to the users’ preferences. Existing prior art consists of methods of storing for large spaces such as warehouses, etc. which have standardized sizes of luggage / containers. However, it would be desirable to have a solution for storing luggage in smaller spaces such as passenger vehicles and to consider luggage having variable sizes. So far, it is challenging to suggest luggage arrangement options according to the boot space and rear seat configurations available in a vehicle. It would further be desirable to enable personalization of the storage of luggage according to the number of users and their individual preferences and priorities and to have a solution which allows users to visualize the arrangement of luggage in storage space. A joint approach of personalization, learning and data-driven model development for suggesting the arrangement of luggage in a vehicle is currently a gap. US 2018099751 A1, AU 2004201888A1, EP 1208521 A1, and US 2002014533 A1 are prior art documents disclosing systems and methods implemented for storing and sequencing of luggage items, methods for identification and measurement of packages by using laser or LADAR (Laser Detecting and Ranging) or holography in a scanning tunnel, and the usage of a camera to determine a size of an object. SUMMARY OF THE INVENTION The object of the invention is to simplify arranging a plurality of items in a stowing space. The object of the invention is achieved by the subject-matter of the independent claims. Advantageous embodiments of the invention are subject-matter of the dependent claims. In one aspect, the invention provides a computer-implemented method for determining an optimized item arrangement in a stowing space by using a computing module, the item arrangement including one or more items to be arranged in the stowing space, the method comprising: a) Collecting and / or receiving information concerning the stowing space and / or the one or more items; and b) Feeding input data based on the information into an item arrangement optimization module that is included in the computing module for determining the optimized item arrangement of the one or more items in the stowing space. An advantage of the method may be that an optimized item arrangement is determined based on input data. The input data is based on information indicated by a user or on information collected and / or received by the system. Thus, the user is conveniently guided how to arrange a plurality of items in a stowing space. Preferably, the stowing space is configured as a loading or boot space of a vehicle, as a shipping container for shipment or as a stowing compartment in a building structure. In embodiments, the items may be configured as luggage items to be arranged in the loading or boot space of the vehicle. Preferably, step b) further comprises: b1) Determining the optimized item arrangement with respect to item dimensions, item weight, item material, item content, item fragility, item criticality, frequency of usage of the one or more items, stowing space dimensions and / or with respect to safety criteria. Preferably, step a) further comprises one, several or all of the following: a1) Collecting and / or receiving item information in an item database; and / or a2) Collecting and / or receiving user profile information of one or more users of the stowing space in a user profile database; and / or a3) Collecting and / or receiving information of a vehicle including the stowing space as loading or boot space, of a planned ride of the vehicle for transportation of the one or more items, and / or of one or more users of the vehicle. An advantage of the method may be that once the system may identify an item and / or one or more users of the stowing space, it can store related item information and / or user profile information. The stored information can be used for future item arrangement optimizations so that the one or more users may need to indicate relevant information only once. This simplifies the usage of the system. The users of the vehicle may be a driver of the vehicle or any other occupant of the vehicle. Preferably, the method further comprises: c1) Collecting and / or receiving image data concerning the one or more items; and c2) Feeding the image data into an item image analyzing module that is included in the computing module for identifying the one or more items in an item database and / or for obtaining item information of the one or more items from and / or for said database. An advantage of the method may be that image data concerning the item can be collected and / or received. Thus, the system can estimate dimensions of the item automatically. Thus, the relevant information may be collected and / or received automatically. This simplifies the usage of the system. Preferably, the method further comprises: d1) Collecting and / or receiving identification data concerning one or more users; and d2) Feeding the identification data into a user identification module that is included in the computing module for identifying the one or more users in a user profile database and / or for obtaining user profile information of the one or more users from and / or for said database. An advantage of the method may be that identification data concerning the one or more users can be collected and / or received. Thus, the system may identify the one or more users and may load relevant personal information, preferences, and priorities automatically. Thus, the relevant information may be collected and / or received automatically. This simplifies the usage of the system. Preferably, the method further comprises: e) Visualizing the optimized item arrangement on a display, preferably of a mobile device and / or of a vehicle, more preferably by using augmented reality techniques. An advantage of the method may be that the determined optimized item arrangement can be visualized by using augmented reality techniques. In other words, the optimized item arrangement may be overlayed with an image of the stowing space displayed on a mobile device or elsewhere. The one or more users are in this way optimally guided how to arrange the one or more items in the stowing space. Preferably, the method comprises: f) Collecting and / or receiving user feedback data concerning the optimized item arrangement and: f1) Feeding input data based on the information and the user feedback data into the item arrangement optimization module for determining an adapted optimized item arrangement; and / or f2) Adapting the item arrangement optimization module based on the user feedback data; and / or f3) Updating the information by using the user feedback data, preferably updating user profile information in a user profile database and / or item information in an item database. An advantage of the method may be that the one or more users can give feedback. The feedback can be used to propose an alternative optimized item arrangement. Furthermore, the feedback may be used to train the computing module. Preferably, the computing module includes at least one of a machine learning model, a neural network model, a regression model, an analytical model, a search algorithm, a linear model, a non-linear model, an evolutionary algorithm, a genetic algorithm, a Bayesian network model, and / or the like. The computing module may be based on and / or derived from algorithms described in the documents and / or websites as referenced herein, see [1] to [8], In another aspect, the invention provides a data processing device comprising means for carrying out the method of any of the preceding embodiments. In another aspect, the invention provides system, preferably a mobile device and / or a vehicle, comprising the data processing device. Preferably, the system comprises an image and / or identification data capturing device for capturing image and / or identification data. In another aspect, the invention provides a computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method of any of the preceding embodiments. In another aspect, the invention provides a computer-readable data carrier having stored thereon the computer program. Aspects of preferred embodiments of the invention may be summarized as follows: In order to provide a comfortable travel experience and maximize storage in a boot space of a vehicle, the proposed system may suggest how to optimally arrange luggage according to personal preferences of users. The system may use inputs from cameras, vehicle data (boot space, rear seat configurations like 60:40 split or full bench, etc.), route information (start and end location, terrain, duration of travel). Artificial intelligence based neural network models such as support vector machines, decision trees, ensemble models, k-nearest neighbor models, Bayesian networks, or other types of models including linear models and / or non-linear models may be used. Neural networks may also include feed-forward neural networks, convolutional neural networks (CNN), recurrent neural networks (RNN). Algorithms used in preferred embodiments of the invention are disclosed and described in the following documents or on the following websites: [1] Kaiming He et al.: “Deep Residual Learning for Image Recognition”; [2] Christian Szegedy et al.: “Rethinking the Inception Architecture for Computer Vision”; https: / / doi.Org / W.48550 / arXiv.1512.00567; [3] Karen Simonyan and Andrew Zisserman: “Very Deep Convolutional Networks for Large-Scale Image Recognition”; https: / / doi.orq / 10.48550 / arXiv.1409.1556; [4] Jorge Lopez Puga et al.: “Bayesian networks”; https: / / doi.org / 10.1038 / nmeth.3550; [5] Feng Liu and Xiaming Liu: “Voxel-based 3D Detection and Reconstruction of Multiple Objects from a Single Image”; htips: / / dai.orq / 10.48550 / arXiv.2111.03098; and [6] Tanweer Alam et al.: “Genetic Algorithm: Reviews, Implementations, and Applications”; https: / / arxiy.^ An idea is to address the issue of arranging luggage in a vehicle for a comfortable travel experience. The proposed system preferably optimizes luggage arrangement in the boot space after personalizing it according to user preferences. Preferred embodiments of the invention use a combination of sensors, inputs, and Al algorithms to personalize and optimize the boot space of a vehicle. Hence, the effort required by an individual in planning and arranging the luggage is reduced, thereby resulting in a better travel experience. The proposed system preferably personalizes luggage arrangement in boot space specifically to user requirements / preferences and historical data. Fig. 3 shows that a mobile device’s front camera is used for user identification and to load a user profile that contains his / her preferences and priorities. If the user is not identified, a guest profile will be created. The biometric modality used for the user identification system can be based on the features available in the vehicle. The user identification and user position detection system can be any biometric system available in the vehicle. This information is used for loading user-specific preferences and boot space personalization in the vehicle. The boot space optimization system (Fig. 4) preferably uses a mobile device’s back camera to capture images of the luggage items. If the quality of the images is high (non-blurry, sharp, or not zoomed-in), the dimensions of the luggage items are estimated. The user provides information about the contents of the luggage such as fragility, criticality, frequency of usage, etc. The luggage dimensions are saved in the mobile app for further usage. Thereby, the proposed system saves time and computation power by re-using the saved luggage dimensions. The virtual dimensions of the luggage may be calculated based on the estimated dimensions from the luggage size estimation or the luggage item analyzing module. The luggage arrangement optimization system may be an Al approach, that can include both statistical ML (machine learning) approaches (decision trees, ensemble models, k-nearest neighbor models, Bayesian networks) and recent deep neural network models to cover multiple aspects of the proposed optimization task. After determining an optimum method of arranging the luggage in the boot space, the visualization engine preferably uses Augmented Reality (AR) on the mobile device’s screen to show the user how the luggage can be arranged according to their preferences in 3D. If the user is not satisfied with the suggested luggage arrangement configuration, another suggestion can also be provided. If the user does not like any of the suggested luggage arrangement configurations, they can manually adjust the luggage as per their preference. In such cases, the luggage arrangement optimization system will preferably take the manual configuration as feedback to provide better suggestions in the future. Embodiments of the invention preferably have the following advantages and effects: - the proposed system preferably uses a mobile device’s camera to capture images. It increases the ease of use, cost-effectiveness, and adaptability of the solution; and / or - since the preferred embodiments use Al algorithms and holistically analyses data such as luggage dimensions, contents, user preferences, vehicle data, etc., it may be much more efficient and mature in making suggestions for arranging luggage in boot space; and / or - the system may also learn how an individual’s preferences and provides better personalized and optimized suggestions for luggage arrangement with regular usage; and / or - since the system may save luggage dimensions in the mobile app for further usage; this saves time and computation power; and / or - the proposed system may increase the security of critical / fragile luggage items by positioning them in such a way that it is not easily accessible to others by understanding their contents and importance; and / or - as augmented reality preferably is used for visualization, it helps the user to better understand how the luggage arrangement should be arranged; and / or - augmented reality-based visualization also allows the users to try their own luggage configuration on the mobile screen and ensure the optimum fit before making the arrangement manually; and / or - ultimately, the user stays comfortable, saves time, and has a better travel experience as they did not have to face the trouble of trying different ways to arrange luggage. BRIEF DESCRIPTION OF THE DRAWINGS Embodiments of the invention are now explained in more detail with reference to the accompanying drawings of which Fig. 1 shows an item arrangement in a stowing space; Fig. 2 shows a first embodiment of a computer-implemented method for determining an optimized item arrangement in the stowing space; Fig. 3 shows an embodiment of a step of the method according to Fig. 2; and Fig. 4 shows a second embodiment of the method. DETAILED DESCRIPTION OF THE DRAWINGS Fig. 1 shows an item arrangement 10 in a stowing space 12. In the case as shown, the stowing space 12 is configured as a loading or boot space of a vehicle 14. The vehicle 14 is in this case a car, but within the scope of the invention, the vehicle 14 may be of any kind such as a land vehicle, aircraft or watercraft. The item arrangement 10 includes a plurality of items 16. A user may arrange the plurality of items 16 in the stowing space 12 depending on different aspects and / or criteria: The item arrangement 10 may depend on item dimensions 18, item content 20, and so on of each item 16. The item arrangement 10 may further depend on stowing space dimensions 22 of the available stowing space 12. In the vehicle 14, the item arrangement 10 may further depend on a rear seat configuration 24 of rear seats and / or a rear bench of the vehicle 14. The item arrangement 10 may further depend on a planned ride, such as a route of the ride, a number of stops, travel times, etc. The item arrangement 10 may further depend on user preferences or priorities 26 of the one or more users during the planned ride, a number 28 of the users or occupants of the vehicle 14 during the ride, etc. An idea is to provide a method for finding the item arrangement 10 that is optimized with respect such criteria. Fig. 2 shows an embodiment of a computer-implemented method for determining the optimized item arrangement 10 in the stowing space 12. The method uses a computing module 30. In a step S11, the method includes: - Collecting and / or receiving information 32 concerning the stowing space 12 and / or the plurality of items 16 to be arranged in the stowing space 12. The step S11 may be implemented on a data processing device comprising means for carrying out the step. Such a data processing device may be a system 39, such a mobile device, smartphone, tablet, and / or the vehicle 14 itself. In step S11, the one or more users may provide information 32 concerning stowing space 12, such as the stowing space dimensions 22, maximum possible load, available fixing means in the stowing space 12, etc. The user may further provide information 32 concerning the items 16, i.e., item information 34, such as the item dimensions 18, item weight, item material, item content, item fragility, item criticality, frequency of usage, etc. The one or more users may further provide information 32 concerning safety criteria. For example, the one or more users may provide information 32 concerning items 16 that should be arranged in a stable or fixed manner or at the bottom of the stowing space 12. If the stowing space 12 is the loading or boot space of a vehicle 14, the one or more users may further provide information 32 concerning the vehicle 14, such as model type, maximum allowed load, etc. In this case, the one or more users may further provide information 32 concerning the planned ride. In the vehicle 14, the one or more users may further provide information 32 concerning the number 28 of users or occupants, etc. The one or more users may further provide information 32 concerning themselves, i.e., user profile information 36, such as the user preferences or priorities 26.The item information 34 and the user profile information 36 may be collected and / or received in an item database 38 and / or a user profile database 40, respectively or commonly. Therefore, this information 32 may be used later for future item arrangement optimizations. The item database 38 and / or the user profile database 40 may be remote databases. The information 32 concerning the stowing space 12 may be collected and / or received in a stowing space database. Therefore, this information 32 may be used later for future item arrangement optimizations. The stowing space database may be a remote database. In a step S12, the method includes: - Feeding input data 41 based on the information 32 into an item arrangement optimization module 42 that is included in the computing module 30 for determining the optimized item arrangement 10 of the plurality of items 16 in the stowing space 12. The step S12 may be implemented on a data processing device comprising means for carrying out the step. Such a data processing device may be the system 39, such a mobile device, smartphone, tablet, and / or the vehicle 14 itself. The item arrangement optimization module 42 may include a machine learning model, a neural network model, a regression model, an analytical model, a search algorithm, a linear model, a non-linear model, an evolutionary algorithm, a genetic algorithm, a Bayesian network model, and / or the like. The item arrangement optimization module 42 is adapted or trained for determining the optimized item arrangement 10. The item arrangement optimization module 42 may be a neural network model trained on real or simulated data. For this, the item arrangement optimization module 42 may be based on evolutionary algorithms such as genetic algorithms, Bayesian network models, or a combination thereof. Furthermore, the item arrangement optimization module 42 may be trained on a dataset having data relating to the input data 41. The optimized item arrangement 10 may be determined with respect to the item dimensions 18, the item weight, the item material, the item content, the item fragility, the item criticality, the frequency of usage of the item, the stowing space dimensions 22 and / or with respect to the safety or other criteria as precedingly mentioned. For example, the optimized item arrangement 10 may indicate to arrange heavy and / or large items 16 at the bottom of the stowing space 12. Items 16 that should be arranged in a stable or fixed manner may be arranged in the vicinity of the fixing means of the stowing space 12 and preferably not in the middle of the item arrangement 10 where such fixing means cannot access. With respect to further security or other criteria, the optimized item arrangement 10 may also indicate to arrange the one or more items 16 in a distributed manner in the stowing space 12, and not piled high only in a portion of the available stowing space 12. If information 32 concerning the vehicle 14, the planned ride and / or the one or more users is collected and / or received in step S11, this information 32 may also be fed as input data 41 into the item arrangement optimization module 42. In a step S13, the method includes: - Collecting and / or receiving image data 44 concerning the one or more items 16; and - Feeding the image data 44 into an item image analyzing module 46 that is included in the computing module 30 for identifying the one or more items 16 in the item database 38 and / or for obtaining the item information 34 of the one or more items 16 from and / or for said database 38. The step S13 may be implemented on a data processing device comprising means for carrying out the step. Such a data processing device may be the system 39, such a mobile device, smartphone, tablet, and / or the vehicle 14 itself. The system 39 may include an image and / or identification data capturing device 52, such as a camera device, for generating the image data 44. For example, the user may capture an image of the item 16 by means of the camera device. The item image analyzing module 46 may include a machine learning model, a neural network model, a regression model, an analytical model, a search algorithm, a linear model, a non-linear model, an evolutionary algorithm, a genetic algorithm, a Bayesian network model, and / or the like. The item image analyzing module 46 is adapted or trained for determining the item dimensions 18, such as length, breadth, height of the item 16. The item image analyzing module 46 may use a neural network model trained on real or simulated data. For this, the item image analyzing module 46 may be based on a CNN, e.g., ResNet, CenterNet-3D, or a combination thereof. Furthermore, the item image analyzing module 46 may be trained on a dataset having images of various items 16. An example of such dataset for luggage items is the dataset “Suitcase / Luggage Dataset Indoor Object Image” which is described and downloadable under: [7] https: / / paperswithcode..com / dataseVsuitc^se4ufl^ In another step (not shown), the method may also include: - Collecting and / or receiving image data 44 concerning the stowing space 12; and - Feeding the image data 44 into a stowing space image analyzing module that is included in the computing module 30 for determining the stowing space dimensions 22 of the available stowing space 12. The latter step may be implemented on a data processing device comprising means for carrying out the step. Such a data processing device may be the system 39, such a mobile device, smartphone, tablet, and / or the vehicle 14 itself. The stowing space image analyzing module may include a machine learning model, a neural network model, a regression model, an analytical model, a search algorithm, a linear model, a non-linear model, an evolutionary algorithm, a genetic algorithm, a Bayesian network model, and / or the like. The stowing space image analyzing module is adapted or trained for determining the stowing space dimensions 22 of the available stowing space 12. The stowing space image analyzing module may a neural network model trained on real or simulated data. For this, the stowing space image analyzing module may be based on a CNN, e.g., ResNet, CenterNet-3D, or a combination thereof. Furthermore, the stowing space image analyzing module may be trained on a dataset having images of various stowing spaces 12. In some embodiments, the deep learning model built to perform estimate dimensions of closed spaces and annotation may be trained on a training dataset comprising between 1000 to 5000 images with labels annotating three to 20 classes of objects with dimensions of each side and coordinates of every corner of the primary object categories. In a step S14, the method includes: - Collecting and / or receiving identification data 48 concerning the one or more users; and - Feeding the identification data 48 into a user identification module 50 that is included in the computing module 30 for identifying the one or more users in the user profile database 40 and / or for obtaining the user profile information 36 of the one or more users from and / or for said database 40. The step S14 may be implemented on a data processing device comprising means for carrying out the step. Such a data processing device may be the system 39, such a mobile device, smartphone, tablet, and / or the vehicle 14 itself. The system 39 may include the image and / or identification data capturing device 52, such as a camera device, for generating the identification data 44. For example, the one or more users may capture an image of themselves by means of the camera device. The user identification module 50 may include a machine learning model, a neural network model, a regression model, an analytical model, a search algorithm, a linear model, a non-linear model, an evolutionary algorithm, a genetic algorithm, a Bayesian network model, and / or the like. The user identification module 50 is adapted or trained for identifying the one or more users. The user identification module 50 may use face recognition based on a neural network model. For this, the user identification module 50 may be based on a CNN, e.g., ResNet, lnceptionV3, or a combination thereof. Furthermore, the user identification module 50 may be trained on a dataset having images containing clearly visible faces. For example, the dataset may contain face photographs of people collected from the web, wherein each face is labelled with the name of the person pictured. An example for such a dataset is “Labeled faces in the Wild” which is described and downloadable under: [81 cs L J Fig. 3 shows an embodiment of the step S14 of the method according to Fig. 2. In a step S21, the identification data 48 is collected and / or received from the image and / or identification data capturing device 52. In a step S22, the identification data 48 is fed into the user identification module 50 for identifying the one or more users in the user profile database 40. In a step S23, the user identification module 50 determines whether there exists user profile information 36 of the one or more users in the user profile database 40, i.e., whether the one or more users can be identified and are known or cannot be identified and are unknown. If the one or more users are unknown, the system 39 and / or the one or more users may generate a new user profile information 36 in a step S24. The user identification module 50 can obtain the user profile information 36 relating to the one or more users. If the one or more users are known, the system 39 and / or the one or more users may load the user profile information 36, particularly the user preferences 26 relating to the one or more users in a step S25. In a step S26, the step S12 is performed. In other words, input data 41 based on the user profile information 36 and the user preferences 26 are fed into the item arrangement optimization module 42. Fig. 4 shows a second embodiment of the method. In a step S31, the step S13 is performed. In other words, image data 44 concerning the one or more items 16 is collected and / or received. Therefore, the one or more users may generate image data 44 by means of the image and / or identification data capturing device 52 in a step S32. For example, the one or more users may capture an image of the item 16 by means of the camera device. In a step S33, the system 39 verifies a quality of the image data 44. If the quality is not enough and too low, the system 39 requests the one or more users to generate new image data 44. If the quality is enough, the image data 44 is fed into the item image analyzing module 46. In a step S34, the item image analyzing module 46 determines whether there exists item information 34 in the item database 38, i.e., whether the item 16 can be identified and is known or cannot be identified and is unknown. If the item 16 is unknown, the system 39 and / or the one or more users may generate a new item information 34. The item image analyzing module 46 can obtain the item information 34 relating to the item 16. If the item 16 is known, the system 39 and / or the one or more users may load the item information 34, particularly the item dimensions 18 and the item content 20 relating to the item 16. The item information 34, in particular, the item dimensions 18 and the item content 20, the user profile information 36, in particular, the user preferences or priorities 26 are then fed into the item arrangement optimization module 42. In case of the vehicle 14, the one or more users may indicate the configuration 24 of the rear seats and / or the rear bench of the vehicle 14, the number 28 of users or occupants of the vehicle 14, the planned ride, or any further information concerning the vehicle 14. The one or more users may further indicate the stowing space dimensions 22 of the available stowing space 12, and / or any further information 32 concerning the stowing space 12 and / or the plurality of items 16. In a step S35, the item arrangement module 42 determines the optimized item arrangement 10. Additionally, the optimized item arrangement 10 may be visualized on a display. In a step S36, the system 39 may collect and / or receive user feedback data concerning the proposed optimized item arrangement 10. In other words, the one or more users may decide whether the proposed optimized item arrangement 10 is acceptable or not. If the proposed optimized item arrangement 10 is acceptable, the method ends. If the proposed optimized item arrangement 10 is not acceptable, the system 39 may, in a step S37, feed input data 41 based on the information 32 and the user feedback data into the item arrangement optimization module 42 for determining an adapted optimized item arrangement 10. Alternatively, if the proposed optimized item arrangement 10 is not acceptable, the system 39 may, in a step S38, adapt the item arrangement module 42 based on the user feedback data, preferably by self-learning. The one or more users may therefore indicate the preferred optimized item arrangement 10 manually. Additionally, or alternatively, the information 32 may be updated based on the user feedback data. In particular, the user profile information 36 in the user profile database 40 and / or the item information 34 in the item database 38 may be updated. In embodiments of the method, the optimized item arrangement 10 may be visualized on a display, in particular in step S35. For example, the optimized item arrangement 10 may be visualized on a display of the system 39. In preferred embodiments, the optimized item arrangement 10 may be visualized by using augmented reality techniques. In other words, the optimized item arrangement 10 may be overlayed with the image data 44 generated by the image and / or identification data capturing device 52, such as the camera device. The invention also provides the data processing device (not shown), the system 39, a computer program (not shown) comprising instructions which, when the program is executed by a computer, cause the computer to carry out the described method, and a computer-readable data carrier (not shown) having stored thereon the computer program. REFERENCE SIGNS 10 item arrangement 12 stowing space 14 vehicle 16 item 18 item dimensions 20 item content 22 stowing space dimensions 24 rear seat configuration 26 user preferences or priorities 28 user or occupant number 30 computing module 32 information 34 item information 36 user profile information 38 item database 39 system 40 user profile database 41 input data 42 item arrangement optimization module 44 image data 46 item image analyzing module 48 identification data 50 user identification module 52 image and / or identification data capturing device

Claims

1. A computer-implemented method for determining an optimized item arrangement (10) in a stowing space (12) by using a computing module (30), the item arrangement (10) including one or more items (16) to be arranged in the stowing space (12), the method comprising:a) Collecting and / or receiving information (32) concerning the stowing space (12) and / or the one or more items (16); andb) Feeding input data (41) based on the information (32) into an item arrangement optimization module (42) that is included in the computing module (30) for determining the optimized item arrangement (10) of the one or more items (16) in the stowing space (12).

2. The method according to claim 1, characterized in that the stowing space (12) is configured as a loading or boot space of a vehicle (14), as a shipping container for shipment or as a stowing compartment in a building structure.

3. The method according to any of the preceding claims, characterized in that step b) further comprises:b1) Determining the optimized item arrangement (10) with respect to item dimensions (18), item weight, item material, item content, item fragility, item criticality, frequency of usage of the one or more items (16), stowing space dimensions (22) and / or with respect to safety or other criteria.

4. The method according to any of the preceding claims, characterized in that step a) further comprises one, several or all of the following:a1) Collecting and / or receiving item information (34) in an item database (38); and / ora2) Collecting and / or receiving user profile information (36) of one or more users of the stowing space (12) in a user profile database (40); and / ora3) Collecting and / or receiving information (32) of a vehicle (14) including the stowing space (12) as loading or boot space, of a planned ride of the vehicle (14) for transportation of the one or more items (16), and / or of one or more users of the vehicle (14).

5. The method according to any of the preceding claims, further comprising:c1) Collecting and / or receiving image data (44) concerning the one or more items (16); andc2) Feeding the image data (44) into an item image analyzing module (46) that is included in the computing module (30) for identifying the one or more items (16) in an item database (38) and / or for obtaining item information (34) of the one or more items (16) from and / or for said database (38).

6. The method according to any of the preceding claims, further comprising: d1) Collecting and / or receiving identification data (48) concerning one or more users of the stowing space (12); andd2) Feeding the identification data (48) into a user identification module (50) that is included in the computing module (30) for identifying the one or more users in a user profile database (40) and / or for obtaining user profile information (36) of the one or more users from and / or for said database (40).

7. The method according to any of the preceding claims, further comprising:e) Visualizing the optimized item arrangement (10) on a display, preferably of a mobile device and / or a vehicle (14), more preferably by using augmented reality techniques.

8. The method according to any of the preceding claims, further comprising:f) Collecting and / or receiving user feedback data concerning the optimized item arrangement (10) and:f1) Feeding input data (41) based on the information (32) and the user feedback data into the item arrangement optimization module (42) for determining an adapted optimized item arrangement (10); and / orf2) Adapting the item arrangement optimization module (42) based on the user feedback data; and / orf3) Updating the information (32) by using the user feedback data, preferably updating user profile information (36) in a user profile database (40) and / or item information (34) in an item database (38).

9. The method according to any of the preceding claims, characterized in that the computing module (30) includes at least one of a machine learning model, a neural network model, a regression model, an analytical model, a search algorithm, a linear model, a non-linear model, an evolutionary algorithm, a genetic algorithm, a Bayesian network model, and / or the like.

10. A data processing device comprising means for carrying out the method of any of the preceding claims.

11. A system (39), preferably a mobile device and / or a vehicle, comprising the data processing device according to claim 10.

12. The system (39) according to claim 9 comprising an image and / or identification data capturing device (52) for capturing image and / or identification data (44, 48).

13. A computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method of any of the claims 1 to 9.

14. A computer-readable data carrier having stored thereon the computer program of claim 13.

Images