CARISSMA
Refine
Document Type
- Conference Paper (138)
- Article (40)
- Part of a Compilation / Book Chapter (9)
- Part of Periodical (5)
- Book (3)
- Preprint (1)
Institute
Review
- peer-review (111)
- nein (37)
- ja (30)
- editorial review (1)
Version
- published (63)
A Novel Approach for Researching Crossing Behavior and Risk Acceptance: The Pedestrian Simulator
(2016)
Normalization of Micro-Doppler Spectra for Cyclists Using High-Resolution Projection Technique
(2019)
Impact scenarios for collisions with Unmanned Aerial Vehicles and their consequences to rotorcraft
(2018)
First Person Trolley Problem: Evaluation of Drivers' Ethical Decisions in a Driving Simulator
(2016)
VRoad: gesture-based interaction between pedestrians and automated vehicles in virtual reality
(2019)
A Bermuda Triangle?: A Review of Method Application and Triangulation in User Experience Evaluation
(2018)
Drowsiness Detection and Warning in Manual and Automated Driving: Results from Subjective Evaluation
(2018)
Who is Generation A? Investigating the Experience of Automated Driving for Different Age Groups
(2018)
Machine Learning Architectures for the Estimation of Predicted Occupancy Grids in Road Traffic
(2018)
This paper introduces a novel machine learning architecture for an efficient estimation of the probabilistic space-time representation of complex traffic scenarios. A detailed representation of the future traffic scenario is of significant importance for autonomous driving and for all active safety systems. In order to predict the future spacetime representation of the traffic scenario, first the type of traffic scenario is identified and then the machine learning algorithm maps the current state of the scenario to possible future states. The input to the machine learning algorithms is the current state representation of a traffic scenario, termed as the Augmented Occupancy Grid (AOG). The output is the probabilistic space-time representation which includes uncertainties regarding the behaviour of the traffic participants and is termed as the Predicted Occupancy Grid (POG). The novel architecture consists of two Stacked Denoising Autoencoders (SDAs) and a set of Random Forests. It is then compared with the other two existing architectures that comprise of SDAs and DeconvNet. The architectures are validated with the help of simulations and the comparisons are made both in terms of accuracy and computational time. Also, a brief overview on the applications of POGs in the field of active safety is presented.
In road traffic, critical situations pass by as quickly as they appear. Within the blink of an eye, one has to come to a decision, which can make the difference between a low severity, high severity or fatal crash. Because time is important, a machine learning driven Crash Severity Predictor (CSP) is presented which provides the estimated crash severity distribution of an imminent crash in less than 0.2ms. This is 63⋅ 103 times faster compared to predicting the same distribution through computationally expensive numerical simulations. With the proposed method, even very complex crash data, like the results of Finite Element Method (FEM) simulations, can be made available ahead of a collision. Knowledge, which can be used to prepare occupants and vehicle to an imminent crash, activate and adjust safety measures like airbags or belt tensioners before of a collision or let self-driving vehicles go for the maneuver with the lowest crash severity. Using a real-world crash test it is shown that significant safety potential is left unused if instead of the CSP-proposed driving maneuver, no or the wrong actions are taken.
A Hybrid Machine Learning Approach for Planning Safe Trajectories in Complex Traffic-Scenarios
(2016)
Simulation-Based Evaluation of ETSI ITS-G5 and Cellular-VCS in a Real-World Road Traffic Scenario
(2018)
The effect of decentralized congestion control on collective perception in dense traffic scenarios
(2018)
A fast airplane boarding strategy using online seat assignment based on passenger classification
(2016)
In 1897 Wilhelm Peukert tested lead-acid batteries with constant current and observed that a single equation can describe the relationship between the discharge capacity of the battery and a constant discharge current. In this article the dependence of the discharge capacity of lithium-ion battery cells, electrochemical double-layer capacitors and lithium capacitors are investigated from low to very high discharge rates. From low to intermediate discharge rates, these energy storage devices show ideal Peukert behavior, but a deviation was observed at high discharge rates. The cells provide less charge than predicted by Peukert’s Law. To describe this deviation, a new equation has been derived by expanding Peukert’s law to very discharge rates. It is capable to describe the discharge behavior of lithium-ion battery cells, electrochemical double-layer capacitors and lithium capacitors from low to high discharge rates in an unequivocal way.
CONTEXT Many universities and vocational training institutions conduct laboratories as simulated experiments. This is due to the costs and supervision needs to conduct hands-on labs safely. Numerous studies have presented mixed opinions on whether hands-on laboratory work is more conducive to learning than a simulated laboratory. Most of the studies put students from experimental and control groups in significantly different conditions. Therefore, it is hard to reach any definite conclusion regarding the influence of the learning mode onto the learning achievements.
PURPOSE This study compares learning outcomes of student laboratory work in an energy storages course conducted in two different modes: first as a practical hands-on exercise and second using computer-based simulations.
APPROACH In order to provide reliable insights, this study implements optimized research methodology to avoid any other effect (e.g. learning synchronicity/distance learning/instructions) on the learning outcome rather than the effect of the learning mode itself. The student laboratory experiments were created in a manner that they could be conducted in both modes in the same way and using a single set of instructions. To ensure a comparable group environment for the individual student, the students were arranged into two similar groups based on the student's practical experience. In this crossover study, the groups were taught the same topics by means of interchanging learning modes.
RESULTS To evaluate the influence of each mode on student learning, short written tests regarding the previous experiment were conducted at the beginning of the subsequent laboratory session. 102 students have taken part in the study in two years. Overall learning results of hands-on experiments were slightly better than those of simulated laboratories (Cohen's d=0.25), the difference in performance was statistically significant (p<0.02). Through solicited feedback on each laboratory session, in hands-on mode more students expressed they have acquired new insights/comprehensions (76% vs. 66%, Cohen's d=0.23, small effect, p<0.07).
CONCLUSIONS Following the strategy not to optimize the lessons individually to the learning mode, other influences on the learning outcome, which were usually mixed, were excluded. The students' subjective opinions show advantages of the hands-on mode. Based on the objective data, a weak, but significant outcome to better knowledge acquisition with hands-on laboratory experiments was achieved. This observation is against the trend of the literature in the last years towards better or equal learning with nontraditional labs. Some of the excluded factors might have a stronger influence on student learning than estimated previously. To get a clear view, the authors recommend isolated research.