Automotive Intelligence for/at Connected Shared Mobility

Project Information

AI4CSM

Grant agreement ID: 101007326

DOI

10.3030/101007326

Project closed

EC signature date 12 May 2021

Start date 1 May 2021

End date 28 February 2025

Funded under

• INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT)

Total cost

€ 41 429 981,73

EU contribution

€ 11 786 405,47

Coordinated by

INFINEON TECHNOLOGIES AG
Germany

Periodic Reporting for period 3 - AI4CSM (Automotive Intelligence for/at Connected Shared Mobility)

Reporting period: 2023-05-01 to 2025-02-28

Summary of the context and overall objectives of the project

Reaching a climate neutral European economy by 2050 is feasible from technological, economic and social perspectives, but it requires the implementation of deep societal and economic transformations throughout the next generation. The major challenge for Europe's industry is to be competitive with the worldwide leading manufacturers in terms of intrinsically robust, scalable and standardized EVs (Electric Vehicleσ) and AVs (Automated Vehicles) for the upcoming mass market based on electronic components and systems (ECS) technologies providing usability for the customer base in Europe.

In this respect AI4CSM aims to enable the future mobility developments following the electrification, standardisation, automatization and digitalisation implementation strategy by providing new AI-enabled electronic component and systems for ECAS vehicles for advanced perception, efficient propulsion and batteries, advanced connectivity, new integration and platform concepts and intelligent components based on trustworthy AI.

Achieving AI4CSM’ s means completing a set of strategic objectives which address the realization of embedded intelligence for a secure connected, cooperative and automated mobility:
- O1 - Develop robust and reliable mobile platforms
- O2 - Develop scalable and embedded intelligence for edge and edge/cloud operation
- O3 - Design silicon for deterministic low latency and build AI-accelerators for decision and learning
- O4 - Solve complexity by trustable AI in functional integrated systems
- O5 - Design functional integrated ECS systems
- O6 - Build ECAS vehicles for the green deal and future connected, shared mobility

AI4CSM positioned Europe as a leader in intelligent, connected, and sustainable mobility technologies. Through deep integration of AI, edge computing, and trusted ECS innovations, it achieved measurable advances in safety, efficiency, and societal trust — laying the foundation for mass-market, climate-neutral electric and autonomous vehicles by 2030.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

• An impressive list of demonstrators is available and has been presented, covering the work n all SCs, leading o the achievement of the objectives, as described above.
• As it would be correct that key impact achievements can be directly linked to the project milestones, a very significant achievement so far has been the completion of the system level design (WP2), concerning all project demonstrators.
• Also, after the end of AI4CSM, the prototypes of components and embedded system developed in WP3 show a high impact towards the functional implementation of the demonstrators envisioned in the supply chains, particularly regarding the acceleration of AI-related and computationally intensive algorithm, functional safe Power Management, enhanced FOD (Foreign Object Detection) capabilities with increased detectability of metallic and non-metallic objects and an inherently redundant gate-driver topology suited for fail-operational behaviour.
• AI4CSM conducts active global communication, promoting high-level scientific and industrial cooperation and the development of more sustainable technologies.
• The AI4CSM demonstrates a significant scientific impact on Mobility related research and development. Numerous publications have been achieved. Public results can be used in further research.
Analyzed topics:
• Performance of Graph Database Management Systems as route planning solutions for different data and usage characteristics
• Towards the Efficient Way of Asphalt Regeneration by Applying Heating and Mechanical Processing
• Frame Error Rate Prediction for Non-Stationary Wireless Vehicular Communication Links
• Extended algorithm for current slope estimation in inverter fed synchronous machines
• Online identification of semiconductor switching times in inverters with inductive load using an FPGA and potential separated comparators
• Multi-objective path tracking control for car-like vehicles with differentially bounded n-smooth output
• Integrated Antenna Module for 5G Applications
• Autonomous Mobile Flock Traffic Simulation in Digital Twin Mode
• Energy-Aware and Fair Multi-User Multi-Task Computation Offloading
• A feasibility study on a traffic supervision system based on 5G communication
• Current Challenges of AI Standardisation in the Digitising Industry
• Ethical Considerations and Trustworthy Industrial AI Systems
• Multivariate outlier explanations using Shapley values and Mahalanobis distances
• An FPGA Overlay for Efficient Real-Time Localization in 1/10th Scale Autonomous Vehicles
• Aspects of foreign object detection in a wireless charging system for electric vehicles using passive inductive sensors as example
• Challenges of testing self-adaptive systems
• Automotive ferrofluidic electromagnetic system for energy harvesting and adaptive damping

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

In AI4CSM, whereas work is reported in WPs, objectives are reached by working in Supply Chains (SCs), each of this works towards reaching one or more project objective and represents a particular research stream aiming to a specific demonstration target. Therefore, the progress beyond SOTA and the results so far are reflected in the work progress within the SCs, as follows:
• SC1 Enables the future mobility developments following the electrification, standardisation, automatization and digitalisation implementation strategy by providing new AI-enabled electronic component and systems for ECAS vehicles for advanced perception, efficient propulsion and batteries, advanced connectivity, new integration and platform concepts and intelligent components based on trustworthy AI.
• SC2 builds an electrical passenger car (EV) to demonstrate AI based fault- detection, analysis, mitigation for the powertrain in real time operation
• SC3 builds coexisting-human operated vehicles and autonomous systems, and the dynamic interaction between them.
• SC4 performs research on and demonstrate a powertrain for xEVs including inverter with AI - based health assessment. AI techniques will be used for AI-based functionalities for the battery management as well as detection of foreign objects by wireless chargers.
• SC5 Performs research on, and demonstrate secure external communication, with high data rates (5G) and bandwidth. The cloud fusion of edge perception results into the digital twin as well as fast and reliable wireless communication channels based on 28 GHz mmW technology
• SC6 Performs research on and demonstrate AI based perception and sensor fusion, as well as new scalable AI-enabled platforms for autonomous mobility interconnected with secure communication architectures and systems.
• SC7 performs research on and demonstrate methods, tools, and processes for a trustable AI-based connected shared mobility with focus of trustworthiness, simulation and virtualization.
• SC8 Implements Europe’s vision of climate neutrality by 2050 for the automotive and the semi-conductor sector. Furthermore, it makes sure that the developments within AI4CSM are conform with current and upcoming standards as well as to support their activates in driving new AI related knowledge into the standards.

overview