Generation and validation of virtual point cloud data for automated driving systems

Hanke, Timo; Schaermann, Alexander; Geiger, Matthias F.; Weiler, Konstantin; Hirsenkorn, Nils; Rauch, Andreas; Schneider, Stefan-Alexander; Biebl, Erwin

doi:10.1109/itsc.2017.8317864

Cited by 36 publications

(59 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For assessment they describe test maneuvers such as sinusoidal or step steering inputs and define KPIs such as response times or overshoot values [91,180]. [79], Schaermann et al [163] apply an overall error, Barons and Pearson correlation coefficients as three validation metrics to validate LIDAR sensor models. They compare multi-dimensional points clouds as raw sensor data as well as occupancy grids as processed sensor data in a parking lot scenario.…”

Section: Deterministic Simulationsmentioning

confidence: 99%

Unified Framework and Survey for Model Verification, Validation and Uncertainty Quantification

Riedmaier

Danquah

Schick

et al. 2020

Arch Computat Methods Eng

View full text Add to dashboard Cite

Simulation is becoming increasingly important in the development, testing and approval process in many areas of engineering, ranging from finite element models to highly complex cyber-physical systems such as autonomous cars. Simulation must be accompanied by model verification, validation and uncertainty quantification (VV&UQ) activities to assess the inherent errors and uncertainties of each simulation model. However, the VV&UQ methods differ greatly between the application areas. In general, a major challenge is the aggregation of uncertainties from calibration and validation experiments to the actual model predictions under new, untested conditions. This is especially relevant due to high extrapolation uncertainties, if the experimental conditions differ strongly from the prediction conditions, or if the output quantities required for prediction cannot be measured during the experiments. In this paper, both the heterogeneous VV&UQ landscape and the challenge of aggregation will be addressed with a novel modular and unified framework to enable credible decision making based on simulation models. This paper contains a comprehensive survey of over 200 literature sources from many application areas and embeds them into the unified framework. In addition, this paper analyzes and compares the VV&UQ methods and the application areas in order to identify strengths and weaknesses and to derive further research directions. The framework thus combines a variety of VV&UQ methods, so that different engineering areas can benefit from new methods and combinations. Finally, this paper presents a procedure to select a suitable method from the framework for the desired application.

show abstract

Section: Deterministic Simulationsmentioning

confidence: 99%

Unified Framework and Survey for Model Verification, Validation and Uncertainty Quantification

Riedmaier

Danquah

Schick

et al. 2020

Arch Computat Methods Eng

View full text Add to dashboard Cite

show abstract

“…Reference [3] shows "a sensor measurement model for an automotive LiDAR sensor". They used raytracing to simulate the sensor measurements, as the working principle of the sensor is very similar with the rendering algorithm.…”

Section: Related Workmentioning

confidence: 99%

“…They used raytracing to simulate the sensor measurements, as the working principle of the sensor is very similar with the rendering algorithm. This "enabled real-time generation of a point cloud within the framework of an automotive simulator" [3].…”

Section: Related Workmentioning

confidence: 99%

“…As mentioned in [3], this method "aims to reproduce the statistical characteristics of errors, i.e. deviations between the perceived and true values" of a sensor system.…”

Section: A Sensor Error Modelmentioning

confidence: 99%

“…Unfortunately, depending on the application and size of desired target, the data acquisition can be very costly and slow [2]. This prompted the development of simulators based on the behavior of 3D imaging sensors for scanning virtual scenes, generating as many synthetic point clouds as the processing capacity of the computer permits [2,3]. This solution is being especially researched for the creation of dataset models for machine learning training in autonomous vehicles [4].…”

mentioning

confidence: 99%

See 2 more Smart Citations

A LiDAR system simulator based on raytracing, modeled with metrological parameters and environmental noise

Gusmão¹,

Barbosa²,

Raposo³

2019

Anais Estendidos Do Simpósio De Realidade Virtual E Aumentada (SVR)

View full text Add to dashboard Cite

The widespread use of point clouds in many applications is prompting many researchers to find new ways to generate such datasets, as the technology is still costly and slow depending on the experiment’s parameters. Virtual scanning, a computer simulation of scanning a real target, is one of these solutions. This paper is a short presentation of the current development of a light detection and ranging (LiDAR) system simulator based on metrological parameters, sensor error model and light scattering treatment.

show abstract

Autonomous Driving Scenario Generation in Overtake Manoeuvres Through Data Fusion

Pinheiro

Stanke

Ferraris

et al. 2020

Mechanisms and Machine Science

View full text Add to dashboard Cite

For an effective study of specific driving scenarios, in particular related to overtaking manoeuvres, developing well-thought-out manoeuvre databases from the acquired data will greatly improve the analysis process. The key point being that identifying clearly the studied scenario and clustering the manoeuvres based on specific sub-cases of this will bring an extra dimension of information that allows to visualise existing correlations between a given set of conditions and the manoeuvres performed under them. This paper expands on how to obtain these databases starting from only from overtaken manoeuvres extracted from experimental data. Consisting in two main procedures, the manoeuvre classification itself, in which a hybrid-classifier that combines both supervised and unsupervised algorithms generates the scenario sub-cases, and the database compilation stage, in where it displays the different types of databases that can be created, based on the aim of the study.

show abstract

Generation and validation of virtual point cloud data for automated driving systems

Cited by 36 publications

References 11 publications

Unified Framework and Survey for Model Verification, Validation and Uncertainty Quantification

Unified Framework and Survey for Model Verification, Validation and Uncertainty Quantification

A LiDAR system simulator based on raytracing, modeled with metrological parameters and environmental noise

Autonomous Driving Scenario Generation in Overtake Manoeuvres Through Data Fusion

Contact Info

Product

Resources

About