This thesis is concerned with automotive active safety, and a central theme is a new safety function called Emergency Lane Assist (ELA). Automotive safety is often categorised into passive and active safety, where passive safety is concerned with reducing the effects of accidents and active safety aims at avoiding them. ELA detects lane departure manoeuvres…
Contents
I Introduction
1 Overview
1.1 Publications
1.2 The scientific contribution of the thesis
2 Trends in automotive active safety
2.1 Active safety functions
2.2 Autonomous vehicles
2.3 Discussion
3 Overview of the papers
3.1 The benefit of active safety functions
3.2 Emergency Lane Assist
3.3 Tracking
3.4 Change detection
3.5 Threat assessment
3.6 Road-aligned coordinates
Bibliography
II Publications
A Towards autonomous collision avoidance by steering 35
1 Introduction
2 Active safety technology evaluation
2.1 Statistics
2.2 Estimating system utility
2.3 Estimating system complexity
2.4 Results
3 Emergency Lane Assist
4 Tracking system
4.1 Motion model
4.2 Measurement model
5 Decision algorithm
6 Evaluation
6.1 Demonstrator vehicle
6.2 Positive performance
6.3 Negative performance
7 Conclusions
References
B Joint road geometry estimation and vehicle tracking
1 Introduction
2 Applications
2.1 Emergency Lane Assist
2.2 Adaptive Cruise Control
2.3 Requirements on the tracking performance
3 Geometric road model
3.1 Overview
3.2 Model derivation
3.3 Coordinate transformation approximations
4 Filtering
4.1 Motion model
4.2 Measurement model
4.3 Decoupled model
4.4 Extended Kalman Filter
5 Evaluation
5.1 Accuracy of the road shape estimate
5.2 Filter tuning
5.3 Mean estimation error
5.4 Lane assignment
5.5 Statistical properties
6 Conclusions
7 Appendix A: Proof of (4) and (10)
References
C Lane departure detection for improved road geometry estimation 89
1 Introduction
2 Estimation Problem
3 Detecting Lane Departures
3.1 Distance Measure
3.2 Stopping Rule
3.3 Application and Result
4 Further enhancement
5 Alternative methods
6 Conclusion
7 Acknowledgment
References
D Statistical threat assessment for general road scenes using Monte Carlo sam-pling
1 Introduction
1.1 Deterministic threat assessment
1.2 Stochastic threat assessment
2 Model .
2.1 Stochastic model
2.2 Dynamic model
2.3 Prior distribution
2.4 Visibility constraints
2.5 Road-aligned coordinates and tracking
3 Threat assessment
4 Implementation
4.1 Collision detection and time discretisation
4.2 Monte Carlo sampling
4.3 Iterative sampling process
5 Results
5.1 Simulated data
5.2 Real data
6 Conclusions
References
E The Marginalised Particle Filter for Automotive Tracking Applications
1 Introduction
2 The combined lane tracking and object model
2.1 Dynamic motion model
2.2 Measurement model
3 Nonlinear estimation
3.1 The Extended Kalman Filter
3.2 The Marginalised Particle Filter
4 Evaluation
4.1 Synthetic data
4.2 Authentic data
5 Conclusions
References
F Obtaining reference road geometry parameters from recorded sensor data
1 Introduction
2 Road model
3 Parameter estimation
3.1 Symmetric sliding..
4 Accuracy of the proposed method
5 Conclusions
Author: Eidehall, Andreas
Source: Linköping University
Download URL 2: Visit Now