AI4Mobile

AI-aided Wireless Systems for Mobility in Industry and Traffic

Learn More

About the Project AI4Mobile

The project AI4Mobile is a research project funded by the Federal Ministry for Education and Research (BMBF), from the announcement
Artificial Intelligence in Communication Networks within the scope of the High-Tech Strategy of the German Federal Government.

Motivation

Powerful and flexible mobile networks are a key technology for numerous branches of industry. Many applications, for example in the fields of automated vehicles or mobile robotics, are highly dynamic and mobile. At the same time, they require extremely reliable communication with low latency. An important challenge is to ensure the required quality of service (e.g. a reliable data rate, no data loss, no delays) for each application. The wireless transmission of information is by nature subject to interference, which means that successful real-time transmission in classic, reactive mobile radio systems cannot be guaranteed. Forward planning of network resources and prediction of relevant quality of service parameters are therefore essential, especially for safety-critical functions. Given the immense complexity of the overall network, methods of artificial intelligence (AI) offer a promising approach here.

Goals and Procedure

The project AI4Mobile aims at developing methods that enable a robust prediction of the quality of service at high mobility using data from the entire mobile network. In combination with other data (application, environment and network data) the predicted quality of service information will be used to develop AI mechanisms for real-time network optimization. AI methods are particularly promising for resource management. In order to make use of existing knowledge in the field of mobile communications, the possibility of supplementing AI approaches with area-specific context information and of integrating expert knowledge into the design of learning processes will be investigated. This will enable a hybrid approach of classical algorithm development and artificial intelligence. The project investigates to what extent this approach can increase the security and robustness of mobile networks.

Innovations and Perspectives

The AI4Mobile solution can be used to predict critical quality of service parameters (e.g. data rate, delay) of a complete communication link between sender and receiver. This allows the implementation of proactive resource optimization and networking in all parts of the mobile infrastructure. An early detection of critical states ensures that the performance of the systems is maintained even if undefined states and faults occur. The security and robustness of the mobile radio network is significantly increased, which is essential for dynamic applications such as automated driving or mobile robotics. The project thus contributes to consolidating Germany's technological leadership, especially in the important automotive and industry 4.0 sectors.

Highlights

Slider
Public Datasets

The consortium has published several datasets on IEEE Dataport, which were collected through ambitious measurement campaigns

Berlin V2X

The Berlin V2X dataset offers high-resolution GPS-located wireless measurements across diverse urban environments in the city of Berlin for both cellular and sidelink radio access technologies, acquired with up to 4 cars over 3 days. The data enables thus a variety of different ML studies towards vehicle-to-anything (V2X) communication.

iV2I+

This measurement campaign was conducted for around 16h at an industrial site where an autonomous cleaning robot collected sensor data from the environment and communicated with a private cellular network.

iV2V

iV2V covers 10h of sidelink communication scenarios between moving and static automated guided vehicles.

All vehicles are equipped with devices that generate background data traffic.

Various features from all parts of the communication system are recorded with a high temporal resolution, resulting in hundreds of megabytes of data.

Software and detailed publications are publicly available for all datasets

Berlin V2X

Github repository
Arxiv preprint

iV2V and iV2I+

Github repository
Arxiv preprint

previous arrow
next arrow
Slider
A9 Measurement Campaign

In October 2020, the first measurement campaign of the project was conducted at the Motorway A9 5G-ConnectedMobility measurement site near Nuremberg, Germany.

Measurements were conducted on five consecutive days with four vehicles travelling a total distance of 3300km.

At the measurement site, five base stations with two sectors each cover motorway, rural, and suburban areas. The non-public network is operated at 700MHz.

Mobile devices in each vehicle use dedicated software to generate data traffic and collect measurements.

Various features from all parts of the communication system are recorded with a high temporal resolution, resulting in several terabytes of data.

All vehicles are equipped with additional mobile devices that generate background data traffic.

previous arrow
next arrow
Slider

Key Figures

Total Cost

7.1
M€

Total Funding

5.1
M€

Funding Rate

71 %

Months

36
03/2020 - 02/2023

Partners

9
Industry & Research

Consortium

Publications

  • D. Wang, A. Qiu, Q. Zhou, S. Partani, and H. D. Schotten, “Investigating the Impact of Variables on Handover Performance in 5G Ultra-Dense Networks,” EuCNC 2023, Gothenburg, Sweden, June 2023.
  • F. Burmeister, A. Palaios, P. Geuer, A. Krause, R. Jacob, P. Schulz and G. Fettweis, “Deep Learning-Supported Kriging for Interpolation of High-Resolution Indoor REMs”, EuCNC 2023, Gothenburg, Sweden, June 2023.
  • P. Geuer, A. Palaios, and R. Zhohov, “Cell- and Area-Based ML Models: Unlocking High Precision Models for Radio Access Networks,” 2023 IEEE Wireless Communications and Networking Conference (WCNC), Glasgow, Scotland, Mar. 2023.
  • M. Parvini, A. Gonzalez, A. Villamil, P. Schulz and G. Fettweis, “Joint Resource Allocation and String-Stable CACC Design with Multi-Agent Reinforcement Learning,” 2023 International Conference on Communications (ICC 2023), Rome, Italy, May 2023
  • W. Donglin, and Hans D. Schotten, “The Effect of Variable TTT, Density, and Velocity on Handover of 5G NR Ultra Dense Network” IEEE Conference on Standards for Communications and Networking (CSCN), Thessaloniki, Greece, 2022.
Cite this work

Privacy

While you are using this website, personal data of you will be processed by us as the person responsible for data processing and stored for the time necessary to fulfil the defined purposes and legal obligations. In the following we will inform you about what data is involved, how it is processed and what rights you have in this respect.

According to Art. 4 No. 1 of the General Data Protection Regulation (GDPR), personal data is any information relating to an identified or identifiable natural person.

This data protection information applies to data processing on all project websites of the project, as well as all services offered in connection with the project as a web service or as a mobile app (for better readability in the following: website), provided that reference is made to this data protection information.

Responsible in the sense of Art. 4 No. 7 GDPR:

Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V.
Hansastraße 27 c
80686 Munich

for its
Heinrich-Hertz-Institute HHI
Einsteinufer 37, 10587 Berlin, Germany

(in the following Fraunhofer-HHI)

E-Mail:datenschutz@zv.fraunhofer.de
Phone:   +49 30 310020
Fax:+49 30 31002-213

The data protection officer of Fraunhofer can be contacted at the above address in Munich, Germany, at Hd. Datenschutzbeauftragter or under datenschutz@zv.fraunhofer.de . If you have any questions about data protection law or your rights as a data subject, you can contact our data protection officer directly at any time.

You can call up our website without having to disclose information about your identity. The browser used on your end device merely automatically sends information to the server of our website (e.g. browser type and version, date and time of access) to enable the website to establish a connection. This also includes the IP address of your requesting terminal device. This is temporarily stored in a so-called log file and deleted after 4 weeks at the latest.

The IP address is processed for technical and administrative purposes of establishing and maintaining the connection, in order to guarantee the security and functionality of our website and, if necessary, to be able to pursue possible illegal attacks on it.

The legal basis for the processing of the IP address is Art. 6 para. 1 sentence 1 lit. f GDPR. Our legitimate interest is based on the security interest mentioned above and the need to ensure that our website is available without disruption.

We cannot draw any direct conclusions about your identity from the processing of the IP address and other information in the log file.

Furthermore, we use cookies when you visit our website. You will find more detailed explanations of this further down in this data protection information.

If we pass on personal data that we collect via the website to processors, we will inform you of this in this data protection information during the respective data processing operation, naming the specific recipient.

Furthermore, we will only pass on your personal data if

  • you have explicitly given your consent pursuant to Art. 6 para. 1 sentence 1 lit. a GDPR
  • this is necessary in accordance with Art. 6 para. 1 sentence 1 lit. b GDPR for the performance of a contract with you (for example, when passing on payment data to shipping companies for the purpose of delivering the goods you have ordered or when passing on payment data to payment service providers or credit institutions in order to carry out a payment transaction)
  • there is a legal obligation for disclosure under Art. 6 para. 1 sentence 1 letter c GDPR.

The data forwarded may be used by the recipients exclusively for the purposes specified.

We use cookies on our website. These are small files that are automatically created by your browser and stored on your end device (laptop, tablet, smartphone or similar) when you visit our site. Cookies do not cause any damage to your terminal device, do not contain viruses, Trojans or other malware.

The cookie contains information that results in connection with the specifically used terminal device. This does not mean, however, that we obtain direct knowledge of your identity.

Der Einsatz von Cookies dient einerseits dazu, die Nutzung unseres Angebots für Sie angenehmer zu gestalten. So setzen wir z.B. sogenannte Session-Cookies ein, um eine SItzungssteuerung zu ermöglichen oder um Formulareingaben oder Warenkörbe während der Sitzung zu speichern. Session-Cookies werden spätestens mit dem Schließen Ihres Webbrowsers gelöscht.

The data processed by cookies is required for the above-mentioned purposes in order to safeguard our legitimate interests and those of third parties in accordance with Art. 6 para. 1 sentence 1 lit. f GDPR.

Most browsers automatically accept cookies. However, you can configure your browser so that no cookies are stored on your computer or so that a message always appears before a new cookie is created. However, disabling cookies completely may mean that you will not be able to use all the features of our website.

You have the right to:

  • revoke your once given consent to us in accordance with Art. 7 Para. 3 GDPR to at any time. As a result, we may no longer continue to process the data which was based on this consent in the future;
  • request information about your personal data processed by us in accordance with Art. 15 GDPR. In particular, you may request information on the purposes of processing, the category of personal data, the categories of recipients to whom your data have been or will be disclosed, the planned storage period, the existence of a right of rectification, erasure, restriction of processing or opposition, the existence of a right of appeal, the origin of your data, if not collected by us, as well as the existence of automated decision making including profiling and, if applicable, meaningful information on the details of such data;
  • immediately request the correction of incorrect or incomplete personal data stored by us in accordance with Art. 16 GDPR;
  • demand the deletion of your personal data stored with us in accordance with Art. 17 GDPR, unless the processing is necessary to exercise the right to freedom of expression and information, to fulfil a legal obligation, for reasons of public interest or to assert, exercise or defend legal claims
  • demand the restriction of the processing of your personal data pursuant to Art. 18 GDPR, if the accuracy of the data is disputed by you, if the processing is unlawful, but you refuse to delete it and we no longer require the data, but you require it for the assertion, exercise or defence of legal claims or if you have lodged an objection to the processing pursuant to Art. 21 GDPR
  • receive your personal data that you have provided us with in a structured, common and machine-readable format or to request its transfer to another responsible party in accordance with Art. 20 GDPR, and
  • complain to a supervisory authority pursuant to Art. 77 GDPR. As a rule, you can turn to the supervisory authority of your usual place of residence or workplace or to the supervisory authority of our company headquarters.

Information about your Right of Objection under Art. 21 GDPR

You have the right to object at any time, for reasons arising from your particular situation, to the processing of personal data relating to you on the basis of Article 6 paragraph 1 letter e) GDPR (data processing in the public interest) and Article 6 paragraph 1 letter f) DPA (data processing based on a balancing of interests); this also applies to profiling of Article 4 No. 4 GDPR based on this provision.

If you object, we will no longer process your personal data unless we can prove compelling reasons for processing that are worthy of protection and outweigh your interests, rights and freedoms, or the processing serves to assert, exercise or defend legal claims.

If you object to the processing of data for the purpose of direct marketing, we will immediately stop processing. In this case it is not necessary to specify a special situation. This also applies to profiling, insofar as it is related to such direct marketing.

If you wish to exercise your right of objection, simply send an e-mail to datenschutzkoordination@zv.fraunhofer.de.

All data that you personally transmit is encrypted using the common and secure TLS (Transport Layer Security) standard. TLS is a secure and proven standard that is also used, for example, in online banking. You can recognise a secure TLS connection by the attached s at http (i.e. https://..) in the address bar of your browser or by the lock symbol at the bottom of your browser.

We also use appropriate technical and organisational security measures to protect your data against accidental or deliberate manipulation, partial or complete loss, destruction or unauthorised access by third parties. Our security measures are continuously improved in line with technological developments.

Due to the further development of our website and offers above or due to changed legal or official requirements it may become necessary to change this data protection information. The current data protection information can be called up and printed out at any time on the website.

Imprint

Heinrich-Hertz-Institute HHI
Einsteinufer 37, 10587 Berlin, Germany

is a legally non-independent institution of

Fraunhofer-Gesellschaft
zur Förderung der angewandten Forschung e.V.
Hansastraße 27 c
80686 Munich
Phone +49 89 1205- 0
Fax +49 89 1205-7531
E-Mail: info(at)zv.fraunhofer.de

www.fraunhofer.de

DE 129515865

Amtsgericht München
Eingetragener Verein
Register-Nr. VR 4461

Dr.-Ing Martin Kasparick
Phone: +49 30 31002-853
E-Mail: martin.kasparick@hhi.fraunhofer.de

Prof. Dr.-Ing. Reimund Neugebauer, President, Corporate Policy and Research Management
Prof. Dr. Ralf Boris Wehrspohn, Technology Marketing and Business Models
Prof. Dr. Alexander Kurz, Human Resources, Legal Affairs and IP Management
Dipl.-Kfm. Andreas Meuer, Finances and Digitalization

Copyright © by
Fraunhofer-Gesellschaft

All rights reserved.
All copyright for this Web site are owned in full by the Fraunhofer-Gesellschaft.

Permission is granted to download or print material published on this site for personal use only. Its use for any other purpose, and in particular its commercial use or distribution, are strictly forbidden in the absence of prior written approval.

Please address your requests for approval to:

Fraunhofer Institute for Telecommunications
Heinrich Hertz Institute
Einsteinufer 37, 10587 Berlin, Germany

Furthermore, you may download or print out a hard copy of this publication solely for the purpose of reporting on the Fraunhofer-Gesellschaft and its institutes, subject to the terms of use set out below.

Graphic alterations to picture motifs - except for cropping the main motif - are not permitted. It is always necessary to quote the source and send two free specimen copies to the above address. The use is free of charge.

 

We assume no liability for the content of external links. The operators of the linked pages are solely responsible for the content of their pages.

We make every effort to ensure that the content of this website is always up-to-date, correct and complete. Nevertheless, the occurrence of errors cannot be completely ruled out. Neither the Fraunhofer Institute nor the Fraunhofer-Gesellschaft accepts liability for the relevance, accuracy or completeness of the information and material offered in its web site. This refers to any material or immaterial damage suffered by third parties as a result of using this Web site.

Protected trademarks and names, images and texts are generally not identified as such on our web pages. However, the absence of such marking does not mean that it is a free name, a free image or a free text in the sense of trademark law.

BMW Group

With its four brands BMW, MINI, Rolls-Royce and BMW Motorrad, the BMW Group is the world’s leading premium manufacturer of automobiles and motorcycles and also provides premium financial and mobility services. The BMW Group contributes to the project with its many years of experience in the areas of complete vehicle technology, alternative drive and energy management concepts, active safety and driver assistance systems as well as information and communication technologies in the vehicle. The strong focus on networked driving is supported by several years of research activities in the field of cellular mobile radio systems for the automotive sector. The BMW Group can draw on extensive project results from national projects (e.g., CONVERGE, 5G NetMobil, Providentia) and EU projects (e.g., preVENT, COMeSafety, METIS).

Robert Bosch GmBH

The BOSCH Group is a leading international technology and service company with the goal of improving people's quality of life through innovative, beneficial and inspiring solutions. The central department Corporate Research and Advanced Engineering bundles the company-wide research activities of BOSCH and brings a lot of experience from previous projects like 5G NetMobil and TACNET 4.0 in the field of communication into the project. In the field of AI research, BOSCH is on the way to becoming a world leader in terms of a wide range of applications. BOSCH participates as a user in the requirements of the use cases, in the technical solutions as well as in the demonstration.

Enway

Enway develops technology for automated and autonomous special vehicles such as sweepers or waste collection vehicles. Several vehicles equipped with Enway technology are already cleaning private areas such as the EUREF Campus in Berlin. First test applications of the technology in German city centres have already been carried out successfully. As a technology provider Enway focuses on the areas of software and machine learning. Together with established vehicle and machine manufacturers, the company works on new, highly automated solutions for cleaning, recycling and waste management.

Ericsson

Ericsson is a leading global supplier of telecommunications infrastructure and services for mobile network operators. Ericsson is also one of the leading companies in the standardization of mobile communications technologies. The research unit at the Ericsson ICT Development Center Eurolab in Herzogenrath close to Aachen specializes in the areas of radio access technologies with extensive experience also in the area of Standardization for 4G and 5G technologies, as well as in exploring solutions in the areas of Industry 4.0 and for automotive use cases. The research group has lead multiple relevant research projects (KOI, 5Gang) and contributed to even more of those (5G NetMobil, ConVeX, 5GCroCo, 5GSMART).

Götting KG

Götting KG, founded in 1965 and based in Lehrte near Hanover, develops and produces radio and sensor technology for commercial applications. The variety of sensors for localization and navigation (lane guidance) of automated guided vehicles (AGVs) and for driver assistance is unique. We are market leaders in many product areas and want to maintain this position in the future. At the same time, we strive to add new products for new markets to our portfolio. Of the current 86 employees, 30 engineers are mainly involved in development.

Safety-critical functions in the vehicle environment in particular can benefit from network-side support, as decisions can be coordinated between the participants and a much larger database is available for decision-making. The decisions made and other data must be made available to the vehicle reliably and in a given time. Also in the area of Industry 4.0, with the increasing use of mobile production robots and other AGVs, these capabilities are crucial for dynamic adaptation of production processes. While mobility and communication processes in the automotive sector are dependent on many factors, communication in the industrial environment is subject to greater regularity and determination, which means that underlying processes can be learned and incorporated into prediction mechanisms.

Fraunhofer Heinrich Hertz Institute

The Fraunhofer Institute for Telecommunications, Heinrich Hertz Institute has been working for about 20 years on the development of modern mobile radio systems in close cooperation with science and industry, resulting in extensive contributions and publications on the theory and practice of resource, network and mobility management in wireless communication networks. In addition, the HHI is active in standardization, among others through the ITU focus group "Machine Learning for Future Networks including 5G". Project-related competences include the following topics: machine learning, kernel methods, compressed sensing, network-supported D2D and multicast communication, coexistence and interference management, network coding and cognitive network management, software-defined networking (SDN), virtualization of networks.

RFmondial

RFmondial has been offering professional products and services in the field of digital radio technologies for more than 10 years. This includes transmitting, receiving and monitoring equipment and related services such as database-based control and evaluation of field measurements. With 12 highly qualified employees a wide range of different technologies is covered. RFmondial takes over the development of a reliable infrastructure for the measurement data as well as the acquisition and pre-processing of training data.

Technische Universität Dresden

The Technische Universität Dresden is one of eleven Excellence Universities in Germany. Its outstanding research strength and success in knowledge and technology transfer have been honored by leading companies with 16 endowed professorships. In the 5G Lab Germany an engineering platform of the TU has been founded in which scientists from 23 professorships work together in interdisciplinary teams to holistically meet the challenges of the fifth generation of mobile communications.

 

The Deutsche Telekom Chair for Communication Networks, led by Prof. Frank Fitzek, provides expertise in the core network, Mobile Edge Cloud (MEC) Computing, SDN, Network Function Virtualization (NFV), smart grids, mesh networks, network slicing and network encoding for reduced end-to-end latency, distributed storage and improved security and resilience. The expertise will be provided by effectively applied the development of 5G to many strategic industries such as robotics (Fast Robotics), energy (National 5G Energy Hub), automotive (5G NetMobil, AutoDrive) and tactile Internet (CeTI, TacNet). angewendet.

 

The Vodafone Chair of Mobile Communications Systems promotes the further development of future wireless technologies through top-level research. Intensive basic research and close cooperation with industry are the key to the success of the projects. The current focus of the chair is on research on the radio level of mobile communications systems, with special emphasis on problems with wireless networks, wireless modems, high frequency interference and implementation architectures of integrated circuits. Of particular interest are problems and solutions that enable an efficient use of radio resources to ensure the service quality of future-relevant applications (tactile Internet, networked automated driving, networked control systems).

Technische Universität Kaiserslautern

TU Kaiserslautern: The research group of Prof. Schotten (Institute for Wireless Communication and Navigation) is a member of the State Research Center "Center for Commercial Vehicle Technology" (ZNT). In the field of mobile and reliable networking solutions, the TUK has already contributed to a large number of projects at European level (e.g. METIS I+II, NORMA) and national level (e.g. HiFlecs, 5G NetMobil, ConVeX). For the evaluation of complex automotive scenarios, simulation-based methodologies are used, which will also be introduced by the partner in to this project. Through the association with "German Research Center for Artificial Intelligence" (DFKI), the partner can also draw on experience from previous projects on AI-based network optimization (including security and reliability aspects as well as self-healing concepts) and incorporate this experience (e.g. EU project Selfnet). This also applies in particular to the design of the AI framework in the project.

en_GBEN
en_GBEN de_DEDE