Pioneering technologies, solutions, new ideas and intelligent concepts, this is the embedded world. We take the opportunity to present ECOMAI at the conference and at the exhibition.
▶️ Free access to the Exhibition, use the code ew25web
▶️ -30% of the Conference Ticket price, use the code EWC25EXH
Meet us at the conference
Booth 4-410
You will meet the following consortium members at the booth:
We sincerly thank the State Development Corporation of Thuringia – LEG for hosting us at the community booth.
Conference: Talks and Tutorial
Tuesday, March 11, 2025
Session 1.1 IOT & CONNECTIVITY, 12:00h – 12:30h
A Visionary Modelling Approach for Predictive Maintenance in a Highly Regulated Environment
Peter Lieber, SparxSystems Europe
Use Case: AI for Platform Screen Door Systems (PSD). PSDs are used in public transportation and separate the waiting area from the rail line, preventing passenger contact with moving vehicles. These systems require high availability (99.4%+), leading to costly preventive maintenance. PdM (Predictive Maintenance) is not yet applied to PSD. This use case discusses the innovation development of Albayrak (Aldoor/Turkiye), a designer and manufacturer of PSD.
Class 7.1: 14:00h – 17:00h
Introduction to tinyML – Running Deep Learning Models on Low-Power Micro-Controllers
Prof. Daniel Mueller-Gritschneder, TU Vienna
Intelligence for the IoT requires to stream huge amounts of data from edge sensors towards the cloud, where deep learning models interpret the data. EdgeAI moves deep learning models from the cloud onto the Edge platforms themselves offering huge gains in terms of connectivity requirements, energy, cost, privacy and end-to-end latency. tinyML or Extreme Edge AI moves the deep learning tasks even further right onto the microcontrollers connected to the sensors.
Thursday, March 13, 2025
Session 7.7 EDGE AI // AI Assisted Motor Control, 12:15h – 12:45h
From Simulation to Silicon on a RISC-V with AI Accelerator
Steven Klotz, Infineon
AI concepts offer sophisticated approaches for condition monitoring and control in embedded applications. However, deploying these strategies on industrial-scale embedded microcontroller systems presents significant challenges. Imperfect simulation models can undermine the successful transfer of developed concepts, limiting their applicability in real systems. Additionally, the limited computational power of real-time microcontrollers necessitates adaptations to ensure efficient execution of learned policies. In this work, a reinforcement learning-based control policy for energy-efficient motor control is presented, highlighting the sim-to-real challenges of deploying this artificial intelligence controller. The AI control concept is studied in the context of a motor control application, with the microcontroller evaluated accordingly. Additional focus is placed on the computational constraints of a low-resource RISC-V microcontroller, presenting toolchain and optimization strategies for real-time operation on cost-effective hardware.