Skip to content
View samlei-research's full-sized avatar
🦇
Not done yet.
🦇
Not done yet.
  • Technical University of Applied Sciences Augsburg
  • Augsburg, Germany

Highlights

  • Pro

Block or report samlei-research

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
samlei-research/README.md

Samuel Leitenmaier

Research Associate at Technical University of Applied Sciences Augsburg

Contact: [email protected]

Biography

Samuel Leitenmaier received his B. Eng. in Computer Engineering and M. Sc. in Applied Research on Computer Science from the Technical University of Applied Sciences Augsburg. He is currently a research associate at the Driverless Mobility research group. His main focus lies on the strong interaction between motion planning and scene understanding of autonomous vehicles with a special focus on heterogeneous computing architectures utilizing these.

Research interests

  • Autonomous Driving
  • Motion Planning, Navigation, Scene Understanding
  • Heterogeneous Computing with FPGAs
  • Efficient Hardware / Software using SoC

Publications

  • Towards Specialized Hardware for Autonomous Driving Functions: THAccelerated Motion Planner. IEEE International Conference on Robotics and Automation , Workshop RoboARCH: Robotics Acceleration with Computing Hardware and Systems, 2025. Abstract
  • Autonomous Electric Race Car Inverter Development: Revving up the future with resource efficient drive technology. IEEE Electrification Magazine, vol. 11, no. 2, 2023. Article

Projects

THAMP: THAccelerated Motion Planner

This project developed as part of the Driverless Mobility research group at Technical Univeristy of Applied Sciences Augsburg.

Results

These results are part of HiL testing and real-world testing with a research vehicle. Four scenarios can be seen.

Straight Driving

HiL Real-world

Parking Vehicle

HiL Real-world

Overtaking

HiL Real-world

Avoidance

HiL Real-world

Yolov4 Edge Object Detection: Jetson TX2 vs. Utlrascale MPSoC

Metrics are: FPS, W. On Jetson TX2 jtop was used to measure power consumption, on Ultrascale MPSoC sysmonwas used.

Detected cars on THA campus using FPGA

Evaluation Hardware

Jetson TX2 UltraZed

Latency

FPS Jetson TX2 UltraZed
Min 3.1 3.091
Max 3.4 3.202
Avg 3.3 3.134

Inference

Power consumption [W] Jetson TX2 UltraZed
Inference Off 2.216 1.734
Inference On 9.955 2.641
Diff. 7.739 0.906

Formula Student Traction Inverter

Hardware Interfaces

Software Interfaces:

Inverter Operating System

Pinned Loading

  1. endat_interface endat_interface Public

    Forked from texane/absenc

    Absolute encoder VHDL core

    VHDL 1

  2. inverter_adc_interface inverter_adc_interface Public

    VHDL

  3. automatic_can_axi_mapper automatic_can_axi_mapper Public

    Bare Metal Software Interface thats running on the mathworks Zynq Linux OS

    C++