AI4CSM

💫This month AI4SCM SC3 partners promise to share some exciting project results... Stay tuned for more information! #demonstrators #mobility #autonomousdriving #EUproject

🔊Today we are at ECCE Europe 2024! 💫Join the #conference to listen to AI4CSM partner Jorge Alejandro Pérez Pérez from ZF Group, giving a presentation about the project-related #research - Feasibility Analysis of Parameter-Based Temperature Estimation of pGaN Power HEMTs Under Consideration of Charge Carrier Trapping! 🌐More information about the conference and full programme: https://lnkd.in/dTf7-U_3

🎯 Accurate and robust object detection is critical for #autonomousdriving! ⛈️🌨️🌫️Image-based detectors face difficulties caused by low visibility in adverse weather conditions. Thus, radar-camera fusion is of particular interest but presents challenges in optimally fusing heterogeneous data sources. 💫To approach this issue, we propose two new radar preprocessing techniques to better align radar and camera data! ✅In addition, we introduce a Multi-Task Cross-Modality Attention-Fusion Network (MCAF-Net) for object detection, which includes two new fusion blocks. ✅These allow for exploiting information from the feature maps more comprehensively. The proposed algorithm jointly detects objects and segments free space, which guides the model to focus on the more relevant part of the scene, namely, the occupied space. ✅Our approach outperforms current state-of-the-art radar-camera fusion-based object detectors in the nuScenes dataset and achieves more robust results in adverse weather conditions and nighttime scenarios. 🔗More information about the research ‘Multi-Task Cross-Modality Attention-Fusion for 2D Object Detection’ https://lnkd.in/dPUv9JU3

💫Last Tuesday AI4CSM project partners Jakob Reckenzaun and Gerhard Dorn from Virtual Vehicle Research GmbH and Moritz Schaffenroth and Martin Fischer from Automotive-Team OTH Amberg-Weiden met Dipl.-Ing. Wolfgang Feigl - Head of Department for Transport at City of Graz in Austria – for a reality check regarding #autonomous driving 🚗 in Graz. 🎯The goal was to discuss the project results (AI4CSM #robotaxi, cloud-based ride-sharing, #standards for #mobility data and data exchange like openDRIVE [1], …) and identify potentials and needs for autonomous applications in Graz. 🗺️The city of Graz is currently working on its SUMP (Sustainable Urban Mobility Plan) 2040 [2] and collecting inputs from all kinds of stakeholders for future mobility concepts.   [1]https://lnkd.in/egH4Qek [2]https://lnkd.in/d8wvkFaT

💫AI4CSM demonstrators integration speeds-up! 📸Last week AI4CSM partners: Automotive-Team OTH Amberg-Weiden, Virtual Vehicle Research GmbH Vehicle and AVL were working on the integration of Mercedes #demonstrator #vehicle!

✨Remembering the great experience of participating in #EEAI2023 last year, we save the date for #EEAI2024: European Conference on EDGE AI Technologies and Applications! 📅October 21-23 🗺️Hotel Regina Margherita in Cagliari, Sardinia, Italy! 💡The EEAI Conference aims to provide a European forum for sharing the latest scientific research and industry results using edge AI technologies and applications. Its scope covers the edge AI technology stack following emerging research and innovation from hardware to software, edge AI frameworks, architectures, algorithms, data types and methods to applications. Stay tuned for the updates!

🔊Save the date for #EFECS2024!

▶️Happy to share AI4CSM partner Politecnico di Torino #video demonstrating the methodology used to validate the drowsiness and #sleep onset prediction #algorithms based on wearable smartwatch and onboard RADAR devices. 🚘😴The video shows how a full-motion #driving #simulation was used as a realistic scenario to validate the algorithms. #Cameras and the #RADAR were installed in the cabin. The volunteers were equipped with a full polysomnograph, wore #smartwatches, and were asked to drive for around 45 minutes on a straight 3-lane highway without the presence of other vehicles. The volunteers were instructed not to use the car infotainment system or to perform other activities, to induce sleep onset. 💡After the tests, sleep expert medical doctors scored the drowsiness state and the occurrence of sleep onsets based on the polysomnograph, which includes an electrocardiogram, medical-grade PPG, and electroencephalogram. Thanks to this scoring, it is possible to assess the algorithms' sensitivity, specificity, and accuracy.

💻Great chance to to listen AI4CSM #standardization expert Erwin Schoitsch from AIT Austrian Institute of Technology presentation "Society5.0 - Sustainability as key challenge for Society and Technology" given at the #Society5.0 #conference in Vienna, and representing AI4CSM and other projects like A-IQ Ready, AIMS5.0, and PowerizeD, targeting #Sustainability, #Technology #Development, and Human-Centered approaches. https://lnkd.in/d9WHQd-5

💫Happy to present interesting AI4CSM project-related #research! 😮💨Respiratory monitoring is becoming an interesting research topic in many areas: health care, #automotive, and aviation. One of the most prominent non-contact #technologies is mm-wave radar, which allows real-time thoracic skin mm-movements tracking. 📰In the paper “3D-Printer-Based Test-Bench for Contactless Respiratory Monitoring Systems” authors: Marco Pogliano, Irene Buraioli, Alessandro Sanginario, Danilo Demarchi and Paolo Motto Ros from Politecnico di Torino focused on developing a test-bench to verify the performance of a breathing rate monitoring device based on a mm-wave radar. 💡The innovative element of the setup concerns a 3D printer programmed by an automatic geometric code (G-code) generation tool that converts an ordered sequence of arbitrary temporal-position coordinates into a series of low-level instructions for reproducing the movement. 💡The test-bench has successfully validated a prototype breath rate detector under lifelike signals and stressful conditions. Future work will leverage this system to design new radar-based respiratory monitoring test-benches and validate new systems! 👆Check out the publication for more information. https://lnkd.in/gPZrYnYh