Hello

I'm Armin Niedermueller

developer and researcher

  • Age 35
  • Address Salzburg, Austria
  • E-mail armin.niedermueller@mailbox.org

Hello! My Name is Armin Niedermueller.  I’m a Research Engineer and Programmer.

Professional Skills

Programming
Robotics
Internet of Things
Computer Vision
Machine Learning
Embedded Systems

Programming Languages and Frameworks

C++
C
Python
Java
Linux
Robot Operating System
Point Cloud Library
OpenCV

Work Experience

2018 - now

Salzburg Research - IoT

Researcher

The focus of my work lies in finding new aspects and technologies in collaborative robotics, machine-to-machine communication and computer vision.

2015 - 2018

ÖBB Technische Services

Hard and Software Developer

I developed new electronics for railcars. This consisted of programming microcontrollers and desigining circuit boards with its electronics. Currently around 400 locomotives and railcars in austria are equipped with electronics by me.

2008 - 2015

ÖBB Technische Services

Electrician - Railcar Maintenance

Maintenance of railcar electronics, air-conditioning, cctv and mechanics.

2001 - 2008

Elektro Berger GmbH

Electrician

Installation of new electrical components and comissioning of ethernet networks in buildings.

Education

2020 - 2022

Master of Science - Applied Image and Signal Processing

University of Applied Sciences & Paris Lodron University- Salzburg

2017 - 2020

Bachelor of Science - Computer science and systems management

University of Applied Sciences - Salzburg

2013 - 2017

Engineer - Electrical engineering and computer science

HTBLuVA Salzburg

1993 - 1997

Primary school

Hauptschule Buermoos

Reference Projects

Collision Avoidance in Human-Robot Collaborative Workspaces using multiple low-cost 3D Cameras

Computer Vision Embedded Systems Hardware Robotics Software

Collision Avoidance in Human-Robot Collaborative Workspaces using multiple low-cost 3D Cameras

Computer Vision Embedded Systems Hardware Robotics Software

Project
Collision Avoidance in Human-Robot Collaborative Workspaces using multiple low-cost 3D Cameras
Company
Salzburg Research
Main Task
With the advent of inexpensive 3D sensors and open source robotics software, low-cost HRC-systems are becoming more competitive against expensive and proprietary systems. For such systems to be safe, the cobot must sense obstacles before touching them. Strict technical requirements exist to realize a safe and efficient system to work with, such as low latency, high accuracy and low memory consumption. This bachelor thesis proposes a low-cost and open source prototype to enable collision detection for an industrial cobot inside a HRC-workspace. The 3D images of two ceiling-mounted cameras are converted into a occupancy voxel grid and the latency together with the memory consumption is analyzed during conversion. Besides, different comparisons between real-world objects and their voxel grid counterpart are conducted to find out the accuracy. Since the voxel grid includes voxels from the cobot itself, the cobot can be misinterpreted as an obstacle and can thus be inoperable. Those voxels are removed by a filter, whose parameters are first optimized before measuring its efficiency. At last, the voxel grid together with a cobot 3D model are visualized inside a virtual workspace.
Results
A low latency is reached, and the memory consumption decreases sharply in the tests after optimizing the OctoMap parameters. Altough the measured accuracy is not suited for fine-grain tasks like the gripping of objects with the cobot itself, it is good enough for the detection of obstacles, which is the main goal of this work. Considering those results, together with the filter removing all cobot voxels, the prototype is suited for collision aware path planning. All in all, the goals of this thesis are reached, and the prototype represents a solid foundation for a HRC-system in which humans can safely work together with cobots in close proximity.

Electronics in a subsystem of the ÖBB 1144 locomotive series and testing hardware.

Electronics Embedded Systems Hardware Software

Electronics in a subsystem of the ÖBB 1144 locomotive series and testing hardware.

Electronics Embedded Systems Hardware Software

Project
Development of a timer relay for the electro locomotive model series ÖBB 1144 as well as development of a testbox in order to automatically test those timer relays.
Company
ÖBB Technische Services
Main Task
Objective Target is the development of a new timer relay and a testbox using state of the art microcontrollers and electronic components. As well as, creating all necessary processes, including their documents, for the mass production and testing of the developed electronics according to ISO 9001.
Tasks
Development of the circuit boards using EAGLE® | Composing program code in „C“ with MPLABX® IDE | Production of an aluminum casing for the testbox | Planning and printing of a case for the reference resistors | Simulation of various circuits using EveryCircuit
Results
Currently around 400 locomotives and railcars in austria are equipped with the new electronics. Processes for manufacturing and testing are fully integrated. The electronics are manufactured and tested at an ÖBB workshop. The number of failures already went significally down.

A method to align images of two 3D-cameras in a collaborative human-robot workspace.

Computer Vision Robotics

A method to align images of two 3D-cameras in a collaborative human-robot workspace.

Computer Vision Robotics

Project
3D point cloud registration using iterative closest point algorithm, point cloud library and robot operating system
Company
Salzburg Research
Topic
In smart factories, humans interact and work together with intelligent robots. Visual information is needed for the robot to safely and efficiently interact with humans. To provide an unobstructed view of a working environment, data from multiple cameras have to be merged. First, two Intel D435i 3D-cameras were mounted on the ceiling above such a human-robot collaborative workspace. This project defines a process using Iterative-Closest-Point (ICP) and the required preprocessing to generate a transformation that can be used to align two 3D-Video feeds. It starts by supplying the fundamentals of the ICP, its variants and the applied preprocessing.
Statement
The ICP works best when the two point clouds are already roughly aligned. The setup in this project consists of two 3D-cameras mounted at different positions at the ceiling and observing a human-robot workspace from completely different angles. This project aims at finding a registration method to align the 3D-point-clouds of both cameras.
Research object
The most effective ICP Variants found in literature were tested on a 3D point cloud data set. 3D point clouds were captured, which represent the expected environment in the workspace and uses them to test the suitability of the Standard-ICP, the nonlinear-ICP (NL-ICP) and the generalized ICP (G-ICP). Additionally the impact of the different preprocessing steps is evaluated. Different preprocessing steps, like filtering and downsampling, were used to improve the results. The main goal was a working prototype application which provides a fully aligned 3d-videostream of the whole workspace.
Conclusion and futher work
It was found that both the NL-ICP and the G-ICP work well with our data-set, while the Standard-ICP returns invalid results for some data-sets. For the preprocessing i found that at least a coarse pre-alignment, downsampling of the point cloud and a restriction of data to contain mostly surfaces observed by both cameras are needed to reliably produce usable results.

My Interests

“For me life is continuously being hungry. The meaning of life is not simply to exist, to survive, but to move ahead, to go up, to achieve, to conquer.”

Arnold Schwarzenegger

  • Computer Vision
  • Machine Learning / Neural Networks
  • Linux
  • Programming
  • Sports
  • Gaming
  • Music
  • Travelling
  • Being A Father
  • Tech Gadgets
  • Investing in Stocks and ETF

Availability Calendar

Sorry. I'm not available on those days