Tobias Bellmann

Dr. Tobias Bellmann


Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

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

Representative Publications

View All Publications
14 September 2020 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
Mohammad Reza Chalak QazaniMohammad Reza Chalak Qazani
Mohammad Reza Chalak Qazani
Houshyar AsadiHoushyar Asadi
Houshyar Asadi
TBTobias Bellmann
Tobias Bellmann
SMShady Mohamed
Shady Mohamed
Chee Peng LimChee Peng Lim
Chee Peng Lim
SNSaeid Nahavandi
Saeid Nahavandi
Motion simulation platforms (MSPs) are widely used to generate driving/flying motion sensations for the users. The MSPs have a restricted workspace area due to the dynamical and physical restrictions of the Motion Platforms active joints as well as the physical limitations of its passive joints. The motion cueing algorithm (MCA) is the reproduction of the motion signal including linear accelerations and angular velocities. It aims to simultaneously respect the MSP's workspace limitations and make the same motion feeling for the user as a real vehicle. The Classical washout filter (WF) is a well-known type of MCA. The classical WF is easy to set-up, offers a low computational burden and high functionality but has some major drawbacks such as fixed WF parameters tuned according to worst-case scenarios and no consideration of the human vestibular system. As a result, adaptive WFs were developed to consider the human vestibular system and enhance the efficiency of the method using time-varying filters. The existing adaptive WFs only cogitate the boundaries of the end-effector in the Cartesian coordinate space as a substitute for the active and passive joints limitations, which is MSP's main limiting factor. This conservative assumption reduces the available workspace area of the MSP and increases the motion sensation error for the MSPs user. In this study, a fuzzy logic-based WF is developed, to consider the dynamical and physical boundaries of the active joints as well as the physical boundaries of the passive joints. A genetic algorithm is used to select the membership functions' values of the active and passive joints' boundaries. The model is designed using MATLAB /Simulink and the outcomes demonstrate the efficiency of the proposed method versus existing adaptive WFs.

18 January 2023 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
Houshyar AsadiHoushyar Asadi
Houshyar Asadi
TBTobias Bellmann
Tobias Bellmann
Mohammad Reza Chalak QazaniMohammad Reza Chalak Qazani
Mohammad Reza Chalak Qazani
SMShady Mohamed
Shady Mohamed
Chee Peng LimChee Peng Lim
Chee Peng Lim
SNSaeid Nahavandi
Saeid Nahavandi
The motion cueing algorithm (MCA) is used to reproduce the driving/flying motion sensation of the real land or air vehicles for the users of motion simulator. Highly accurate MCAs are required for the motion simulators to create realistic sensation for the simulator users, otherwise, they might cause motion sickness and user discomfort. Model predictive control (MPC) is a popular technique in development of the high-fidelity MCAs as it is able to consider the motion sensation as well as the workspace constraints of the simulator platform. Tilt coordination in MCA is in charge of creating sustained linear acceleration feeling through tilting the platform cabin and exploiting gravitational acceleration under human motion threshold, to not allow the user to perceive this as a rotational motion. However, the existing MPC-based MCAs have not considered the rate limit to constrain the tilt motion under human threshold which causes violations of the rotational sensation threshold during generating sustained acceleration feeling. Instead, some researchers increase the penalty weights of rotational motions in order to slow down rotational motions. Using this strategy slows down all the simulator rotational motions not only the ones coming from tilt coordination, to generate sustained acceleration feeling, but also from rotational channel (due to real vehicle rotation) which causes motion sensation error and consequently motion sickness. Therefore, as the existing MPC-based MCAs cannot differentiate the rotations of tilt coordination channel (due to sustained acceleration feeling generation), from rotational channel (due to vehicle rotational motions), they are not able to produce accurate rotational motion sensations because of insufficient rotational motions. In this research, a novel decoupled MPC-based MCA is developed to systematically address the issues related to the existing MPC-based MCAs, in terms of inaccurate motion generation, by redesigning the MPC-based MCA using a series of vestibular system-based MPC models and considering tilt rate limit in tilt coordination. The simulation study using MATLAB software is used to verify and validate the proposed MCA compared to the existing MCAs. The proposed decoupled MPC-based MCA is able to generate accurate motion cues compared to those of the existing MCAs.

4 July 2017
Verified
Conference Paper
Open access
PDF file
BTBernhard Thiele
Bernhard Thiele
TBThomas Beutlich
Thomas Beutlich
VWVolker Waurich
Volker Waurich
Martin SjölundMartin Sjölund
Martin Sjölund
TBTobias Bellmann
Tobias Bellmann
Link?ping University Electronic Press Conference Proceedings

18 March 2024 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
HSHendrik Scheidel
Hendrik Scheidel
Houshyar AsadiHoushyar Asadi
Houshyar Asadi
TBTobias Bellmann
Tobias Bellmann
ASAndreas Seefried
Andreas Seefried
SMShady Mohamed
Shady Mohamed
SNSaeid Nahavandi
Saeid Nahavandi
In the field of motion simulation, the level of immersion strongly depends on the motion cueing algorithm (MCA), as it transfers the reference motion of the simulated vehicle to a motion of the motion simulation platform (MSP). The challenge for the MCA is to reproduce the motion perception of a real vehicle driver as accurately as possible without exceeding the limits of the workspace of the MSP in order to provide a realistic virtual driving experience. In case of a large discrepancy between the perceived motion signals and the optical cues, motion sickness may occur with the typical symptoms of nausea, dizziness, headache and fatigue. Existing approaches either produce non-optimal results, e.g. due to filtering, linearization, or simplifications, or the required computational time exceeds the real-time requirements of a closed-loop application. This work presents a new solution to the motion cueing problem, where instead of a human designer specifying the principles of the MCA, an artificial intelligence (AI) learns the optimal motion by trial and error in interaction with the MSP. To achieve this, a well-established deep reinforcement learning (RL) algorithm is applied, where an agent interacts with an environment formulated as a Markov decision process (MDP). This allows the agent to directly control a simulated MSP to obtain feedback on its performance in terms of platform workspace usage and the motion acting on the simulator user. The RL algorithm used is proximal policy optimization (PPO), where the value function and the policy corresponding to the control strategy are learned and both are mapped in artificial neural networks (ANN). This approach is implemented in Python and the functionality is demonstrated by the practical example of pre-recorded lateral maneuvers. The subsequent validation on a standardized double lane change shows that the RL algorithm is able to learn the control strategy and improve the quality of the immersion compared to an established method while enhancing the realistic driving motion sensation. Thereby, both the perceived translational accelerations and rotational angular velocities determined under consideration of the vestibular system are more accurately reproduced, and the resources of the MSP are used more economically.

1 April 2019
Verified
Conference Paper
Closed access
ASAndreas Seefried
Andreas Seefried
APAlexander Pollok
Alexander Pollok
RKRichard Kuchar
Richard Kuchar
MHMatthias Hellerer
Matthias Hellerer
MLMartin Leitner
Martin Leitner
TBTobias Bellmann
Tobias Bellmann