Maikol Drechsler

Maikol Drechsler

Auto-generated profile, based on publicly available information

Maikol Drechsler

CARISSMA Institute of Automated Driving, Technische Hochschule Ingolstadt, Ingolstadt, Germany

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 March 2022 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
MDMaikol Funk Drechsler
Maikol Funk Drechsler
JPJakob Peintner
Jakob Peintner
FRFabio Reway
Fabio Reway
GSGeorg Seifert
Georg Seifert
Andreas RienerAndreas Riener
Andreas Riener
WHWerner Huber
Werner Huber
Scenario-based proving ground testing has been established as the standard test environment for the homologation of driver assistance systems. These test procedures are described, for example, in the New Car Assessment Programme test catalog, which only considers a limited number of possible scenarios. Furthermore, pedestrian targets are, among others, used to test assistance systems for protection against vulnerable road users (VRUs). However, these targets also have the disadvantage that they only represent humans in a simplified way and cannot depict realistic movement patterns. On the other hand, simulation has been established as an efficient means of realizing scenarios in a reproducible, time- and cost-saving manner. However, simulation has a major disadvantage: it can only represent reality to a limited extent, e.g., in the influence of human factors in response to hazardous situations. To bridge the gap between simulation and real-world testing, we introduce MiRE, a mixed reality environment for testing automated driving functions. In MiRE, real entities meet in a virtual environment and take advantage of real and simulated test procedures. This paper presents MiRE, a proof of concept based on a modular architecture. The goal is to use it to evaluate the feasibility of testing in mixed reality – by comparing it to proving ground testing. Preliminary results show that MiRE is a suitable environment for testing complex scenarios where the real vehicle and VRUs can safely interact in safety-critical situations and produce realistic results.

7 December 2021 in Modelling and Simulation in Engineering
Verified
Journal Article
Open access
PDF file
MDM. Funk Drechsler
M. Funk Drechsler
TFThiago Antonio Fiorentin
Thiago Antonio Fiorentin
HGH. Göllinger
H. Göllinger
The use of actor-critic algorithms can improve the controllers currently implemented in automotive applications. This method combines reinforcement learning (RL) and neural networks to achieve the possibility of controlling nonlinear systems with real-time capabilities. Actor-critic algorithms were already applied with success in different controllers including autonomous driving, antilock braking system (ABS), and electronic stability control (ESC). However, in the current researches, virtual environments are implemented for the training process instead of using real plants to obtain the datasets. This limitation is given by trial and error methods implemented for the training process, which generates considerable risks in case the controller directly acts on the real plant. In this way, the present research proposes and evaluates an open-loop training process, which permits the data acquisition without the control interaction and an open-loop training of the neural networks. The performance of the trained controllers is evaluated by a design of experiments (DOE) to understand how it is affected by the generated dataset. The results present a successful application of open-loop training architecture. The controller can maintain the slip ratio under adequate levels during maneuvers on different floors, including grounds that are not applied during the training process. The actor neural network is also able to identify the different floors and change the acceleration profile according to the characteristics of each ground.

20 August 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
PCPak Hung Chan
Pak Hung Chan
ZWZiXiang Wei
ZiXiang Wei
MDMaikol Drechsler
Maikol Drechsler
KDKurt Debattista
Kurt Debattista
VDValentina Donzella
Valentina Donzella
Assisted and automated driving systems rely on perception sensor data to ensure safe operations, especially in safety-critical applications like in the intelligent transportation sector. However, adverse weather is still a disruptive scenario, impacting sensor data and resulting automated functions. For this reason, it is critical to be able to develop, test and validate automated systems under these adverse conditions. Noise augmentation of sensor data can expand the number of test cases without compromising safety. This work presents a set of novel physics-based weather noise models for camera, augmenting real world or synthetic images. These models fully consider the physical properties of the weather phenomena alongside their interactions with visible light and camera parameters. Unlike existing works, these models are intended to be applied on Bayer images, matching where the real effect of the weather will be first captured by a camera, therefore enabling bespoke testing based on the automated system. The effect created by the newly proposed and validated models is compared against real-world paired data, collected in controlled manner, and with different weather severity levels. Augmented data with rain and fog noise followed expected trends for the selected metrics, whilst the snow noise had some variability. Overall the trends of quantitative evaluations are in agreement with real-data, and future work will include more accurate data collection and weather modelling to further improve the models. These models have been deployed in a mixed-reality Vehicle-in-the-Loop testing pipeline, enabled verifiable and controlled testing of the full autonomy system under different degrading conditions.

30 May 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
HHHamed Haghighi
Hamed Haghighi
PSPierpaolo Serio
Pierpaolo Serio
MDMaikol Drechsler
Maikol Drechsler
JRJonathan Robinson
Jonathan Robinson
DGDaniel Gummadi
Daniel Gummadi
VDValentina Donzella
Valentina Donzella
LiDAR sensors play a crucial role in automated driving by delivering precise three-dimensional measurements of the environment. Their reliability, however, can be compromised under adverse weather, e.g., when scattering and attenuation from rain, fog, and snow introduce spurious returns and signal loss. These effects can hinder object detection and tracking; however, it is difficult to evaluate this performance decrease because collecting representative real-world data under such conditions is challenging, given their unpredictability and variability. To address this challenge, this paper introduces three novel high-fidelity LiDAR physics-based adverse-weather models and proposes and demonstrates a methodology to validate them with real-world data. The validation against paired data provides quantitative evidence of model fidelity. The proposed models not only offer a controlled way to augment training data but also enable integration for real-time controlled testing to evaluate automated driving functions under diverse weather conditions in simulation, accelerating the development of safe automated driving systems.