A new method for the objective assessment of ADAS based on multivariate time series classification

Moser, Uwe; Harmening, Nick; Schramm, Dieter

doi:10.1007/978-3-658-25939-6_53

Cited by 4 publications

(7 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, we cannot simply correlate the characteristic values of lateral deviation/velocity/acceleration with "perceived safety" and expect a robust causal relationship of the model. This could be solved with a) more complicated models with more predictors, individually or using factor analysis, which might however be hard to interpret as a result; b) using more detailed subjective criteria on expert-level as targets of the model; c) using other ML algorithms such as pattern recognition [46] or time series classification [68] which preserves more information in the objective data than just the characteristic values. 4) other ML algorithms and manual driving analysis -these models normally require more data for training, but with an extensive database they still have great potentials.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

“…Professional test drivers took part in the driving tests in Ref. [29] [56] [68]. The selection of test drivers also has an influence on the design of the questionnaire or even of the whole procedure.…”

Section: B Various Subjective Assessment Strategiesmentioning

confidence: 99%

“…[46] to cluster the driving data into different styles. Moser et al [68] pointed out the limitations of the state-of-the-art correlation and regression methods, which includes the variance of the subjective perceptions, high impact of the system designers, and the information loss due to the reduction of time series data to a characteristic value. The application of NN like in Ref.…”

Section: A Other Machine Learning Algorithmsmentioning

confidence: 99%

“…The manual driving data were analyzed using ANOVA and Bonferroni-corrected post-hoc tests, in order to identify if the selected KPIs distinguish between different styles. The abovementioned publication [68] also made use of data collected from synthetic oscillation in lane keeping scenario and manual lane change maneuver.…”

Section: B Manual Driving Data Analysismentioning

confidence: 99%

See 3 more Smart Citations

State of the Art: Ongoing Research in Assessment Methods for Lane Keeping Assistance Systems

Wei

Pfeffer

Edelmann

2024

IEEE Trans. Intell. Veh.

View full text Add to dashboard Cite

A safe, comfortable and intelligent Lane Keeping Assistance System (LKAS) is fundamental on the way to realizing full automated driving. In spite of the current high market penetration of LKAS, vehicle manufacturers, suppliers and researchers are still working on improving the system to reach a higher customer acceptance. To achieve this goal more efficiently, a systematic and ideally objective assessment procedure of the system needs to be established. Till date, a thorough review study of the assessment procedures of LKAS does not exist, especially of those which also take the subjective impression of the driver into consideration. The motivation of this paper is to benchmark the state-of-the-art assessment methods of LKAS and identify where problems and research potentials lie. The paper summarizes the relevant characteristics of LKAS based on representative research in the past decades. It also compares the characterization of LKAS with the field of steering feel and vehicle handling in the aspects of procedure, strategy, as well as their advantages and disadvantages. This paper contributes to transferring the know-hows and valuable experiences from other well-developed fields of vehicle dynamics into the rapid-evolving branch of Advanced Driver Assistance System (ADAS), in order to help standardizing its assessment procedure and give insights on system design from a human-centered perspective in the early development phase, thus shortening the ADAS development cycles.

show abstract

Section: Conclusion and Discussionmentioning

confidence: 99%

Section: B Various Subjective Assessment Strategiesmentioning

confidence: 99%

Section: A Other Machine Learning Algorithmsmentioning

confidence: 99%

Section: B Manual Driving Data Analysismentioning

confidence: 99%

See 2 more Smart Citations

State of the Art: Ongoing Research in Assessment Methods for Lane Keeping Assistance Systems

Wei

Pfeffer

Edelmann

2024

IEEE Trans. Intell. Veh.

View full text Add to dashboard Cite

show abstract

“…The scenario catalog consists of six scenarios that are defined as follows: In order to measure the performance for each scenario, objective functions need to be defined. In this work a machine learning based approach is used for an objective assessment of driving behavior presented by Moser et al [38]. This approach uses multivariate time series classification to evaluate driving comfort for lane keeping, lane change and acceleration/deceleration maneuvers.…”

Section: A Problem Descriptionmentioning

confidence: 99%

Efficient Simulation Based Calibration of Automated Driving Functions Based on Sensitivity Based Optimization

Fraikin

Funk

Frey

et al. 2020

IEEE Open J. Intell. Transp. Syst.

View full text Add to dashboard Cite

Comfort and Safety in Conditional Automated Driving in Dependence on Personal Driving Behavior

Vasile,

Dinkha,

Seitz

et al. 2023

IEEE Open J. Intell. Transp. Syst.

Self Cite

View full text Add to dashboard Cite

When changing from active driving to conditional automated driving (CAD), the question arises whether users still prefer their own driving behavior while being a passenger. The aim of this paper is to analyze driving behavior preferences in CAD based on the perception of comfort and safety, taking the personal driving behavior into account. Furthermore, it is investigated if users are able to manually demonstrate their desired driving behavior for CAD. Data on the personal, desired and experienced automated driving behavior of 42 participants from a real-world study was used to investigate both research questions for car-following (CF) and decelerating to a lead vehicle (DL) situations. In a first step, the personal and desired driving behavior is compared with the automated driving behavior based on selected parameters. Subsequently, the relationship between behavior differences and the assessed situation comfort and safety is analyzed. The results show a dependency between differences of personal and automated driving behavior and subjective ratings for comfort and safety. Furthermore, results suggest that participants prefer a driving behavior similar to or more defensive than their own for CAD. Our findings also show that participants were able to demonstrate their desired comfort driving behavior in CF and DL situations.

show abstract

A new method for the objective assessment of ADAS based on multivariate time series classification

Cited by 4 publications

References 11 publications

State of the Art: Ongoing Research in Assessment Methods for Lane Keeping Assistance Systems

State of the Art: Ongoing Research in Assessment Methods for Lane Keeping Assistance Systems

Efficient Simulation Based Calibration of Automated Driving Functions Based on Sensitivity Based Optimization

Comfort and Safety in Conditional Automated Driving in Dependence on Personal Driving Behavior

Contact Info

Product

Resources

About