Tim Andersson

Tim Andersson

Educational background

He first studied maintenance technology regarding airplanes and trains for 4 years, and then worked as a train technician/mechanic for 8 years. He has also done military service as a mechanic on JAS 39 Gripen. This experience has given him an understanding of how different machines work, how they wear over time and be careful while performing his work regardless of stress level. The years of working with maintenance inspired him to be an engineer. In 2015 he started studying for a Master of Science in robotics. The robotics program is an interdisciplinary education that consists of both high- and low-level programming, mechanical- and electrical construction, this makes him very versatile as an engineer.

Work Experience

After another three years of studies, he got a one-year employment contract as a mechanical engineer. He is currently working as an industrial doctoral at Assa Abloy where he is doing research regarding detection and classification of mechanical anomalies in cylinder locks by using torque measurements and machine learning. 

Research Interests

His research is within computer science regarding applied machine learning.


He has been part of two publications during his Master’s degree regarding machine learning.

(2019) A Machine Learning Approach for Biomass Characterization

In this study, an evaluation was done different combinations of pre-processing- and machine learning regression- methods regarding moisture prediction in biomass using near-infrared spectroscopy. It was also evaluated which wavelength combinations that were most important for the regression models for their predictions. My contribution to this study was the evaluation of wavelengths using a genetic algorithm.

(2020) Road Boundary Detection Using Ant Colony Optimization Algorithm

In this study, an algorithm was developed to define coherent road boundaries in video data during night conditions for an unstructured road. An evolutionary approach was used where an algorithm was developed to find the best starting positions for the two ant colonies which were instructed to find and follow the road boundaries and at the same time create coherent white lines representing the boundaries. To achieve this, an autonomous algorithm was developed to adapt the pre- processing of the video data for the changing light conditions. My contribution to this study was the modification of the ant optimization algorithm and the pre-processing algorithm.

Become the next PhD Student!

Apply now for a membership into our Doctoral Student Network and join the IndTech Community!