Jan Strohbeck

Jan Strohbeck

PhD student

Ulm University

Biography

Jan Strohbeck (Graduate Student Member, IEEE) received the Bachelor of Engineering degree in Electrical and Computer Engineering and the Master of Science degree in Computer Science from Aalen University, Germany. Since 2018, he has been working as a Researcher at the Institute of Measurement, Control and Microtechnology, Ulm University, Germany. His research interests include motion prediction of traffic participants, specifically at intersections and the estimation of the prediction’s inherent uncertainty.

Interests
  • Artificial Intelligence
  • Automated driving
  • Trajectory forecasting
Education
  • M.Sc. in Computer Science, 2018

    Aalen University

  • B.Eng. in Electrical and Computer Engineering, 2016

    Aalen University

Experience

 
 
 
 
 
Ulm University, Institute for Measurement, Control and Microtechnology
Research assistant, PhD student
August 2018 – Present Ulm, Germany
  • Research topic: Trajectory forecasting of traffic participants for automated driving
  • 5 first author as well as multiple co-author publications at IEEE conferences (IROS, ITSC, VNC)
  • Participation in EU/BMWK projects MEC-View, ICT4CART, LUKAS, EVENTS
  • 1st place at the 2019 Argoverse Motion Forecasting Challenge
  • Supervision of 6 Master and Bachelor theses
  • Exercises supervisor for lectures “Introduction to Deep Learning”, “Driver assistance systems”
  • Administration of the Gitlab instance of the institute, creation and maintenance of many build tools and CI infrastructure
  • ROS/ROS2, deep learning, neural networks, CNNs, GCNs, Gaussian processes, PyTorch, Python, C++, CUDA, OpenGL, Docker, CMake
 
 
 
 
 
Live Statistics Darts GbR
Cross-platform app for darts tournaments
October 2017 – June 2018 Aalen, Germany
  • Partly as a Master thesis
  • Cross platform darts scorebord app
  • Online player profiles, social feed
  • Creation, organization, playing and management of darts tournaments
  • React Native cross-platform app, React web app with server-side rendering, Feathers.js backend API with MongoDB
  • TypeScript, React Native, React, Node.js, Feathers.js, MongoDB, Docker, Kubernetes, CI/CD (Jenkins), Java, Swift
 
 
 
 
 
Carl Zeiss SMT GmbH
Bachelor thesis at Carl Zeiss SMT GmbH
September 2015 – February 2016 Oberkochen, Germany
  • Creating a software to simulate multiple hardware devices for the purpose of testing the hard- and software of a mirror testing apparatus.
  • Implementation of base classes to simplify the usage of protocols (CAN, CANopen, TCP, Modbus, proprietary protocols)
  • Software-based simulation of hardware responses (e.g., of a mirror-lifting robot)
  • C#, C++, C
 
 
 
 
 
LOBO Electronic
Internship at LOBO Electronic
September 2014 – February 2015 Aalen, Germany
  • Creating a laser beam position regulation system using an Atmel microcontroller, reflective mirrors and rotating mounts.
  • Implementation of a firmware update protocol
  • Qt GUI for showing alignment status/progress and for firmware update
  • C, C++, Qt

Recent Publications

(2023). Graph-based Trajectory Prediction with Cooperative Information. In IEEE ITSC 2023.

PDF Cite DOI

(2022). Deep Kernel Learning for Uncertainty Estimation in Multiple Trajectory Prediction Networks. In IEEE/RSJ IROS 2022.

PDF Cite DOI

(2021). An Extension Proposal for the Collective Perception Service to Avoid Transformation Errors and Include Object Predictions. In IEEE VNC 2021.

PDF Cite DOI

(2021). DeepSIL: A Software-in-the-Loop Framework for Evaluating Motion Planning Schemes Using Multiple Trajectory Prediction Networks. In IEEE/RSJ IROS 2021.

PDF Cite DOI

(2020). Multiple Trajectory Prediction with Deep Temporal and Spatial Convolutional Neural Networks. In IEEE/RSJ IROS 2020.

PDF Cite DOI

Projects

Live Statistics Darts
Innovative darts scoring app

Contact

If you want to get in touch, message me on LinkedIn, use my work phone number, or use the email address given in the imprint.