Wolfgang Branz

Research on trajectory planning in emergency situations with multiple objects

This paper answers the question of the latest possible evasive trajectory. Using a novel approach, it can be shown and proven that it has to be a combined braking and steering maneuver. Although Horiuchi et al. (2004) has already pointed out the advantages of a combination over a sole steering intervention, the question of the latest possible collision-free trajectory has not yet been answered. Based on the novel approach, a propagation method is presented, which reveals unavoidable collisions in scenarios with multiple obstacles and which can be used to derivate an evasive trajectory

Estimating the Process Noise Variance for Vehicle Motion Models

Vehicle motion models are employed in driver assistance systems for tracking and prediction tasks. For probabilistic decision making and uncertainty propagation, the predictions inaccuracy is taken into account in the form of process noise. This work estimates Gaussian process noise models from measured vehicle trajectories using the expectation maximisation (EM) algorithm. The method is exemplified and the results evaluated for three commonly used motion models based on a large-scale dataset. A novel closed-form adaptation of the algorithm to a covariance matrix with Kronecker product structure, as in models for translational motion, is presented. The findings suggest that the longitudinal prediction errors feature a non-Gaussian distribution but a reasonable approximation is given by the estimated model.

Uncertainty propagation in criticality measures for driver assistance

Active safety systems employ surround environment perception in order to detect critical driving situations. Assessing the threat level, e.g. the risk of an imminent collision, is usually based on criticality measures which are calculated from the sensor measurements. However, these metrics are subject to uncertainty. Probabilistic modelling of the uncertainty allows for more informed decision making and the derivation of sensor requirements. This work derives closed-form expressions for probability distributions of criticality measures under both state estimation and prediction uncertainty. The analysis is founded on uncertainty propagation in non-linear motion models. Finding the distribution of model-based criticality metrics is then performed using closed-form expressions for the collision probability and error propagation in implicit functions. All results are illustrated and verified in Monte-Carlo simulations.

Sensor Data Fusion for Active Safety Systems

Active safety systems will have a great impact in the next generation of vehicles. This is partly originated by the increasing consumers interest for safety and partly by new traffic safety laws. Control actions in the vehicle are based on an extensive environment model which contains information about relevant objects in vehicle surroundings. Sensor data fusion integrates measurements from different surround sensors into this environment model. In order to avoid system malfunctions, high reliability in the interpretation of the situation, and therefore in the environment model, is essential. Hence, the main idea of data fusion is to make use of the advantages of using multiple sensors and different technologies in order to fulfill these requirements, which are especially high due to autonomous interventions in vehicle dynamics (e. g. automatic emergency braking). The technical challenge in the development of a serial product relies in the implementation with given sensors, as well as in the risk assessment of the system.

Analytical derivation of performance bounds of autonomous emergency brake systems

Autonomous emergency brake (AEB) systems have to decide on brake interventions based on an uncertain and incomplete perception of the environment. This paper analyses theoretical limitations in AEB systems caused by noisy sensor measurements and uncertain prediction models. Such performance bounds can be used to derive sensor accuracy constraints, to identify challenging scenarios or to develop objective metrics. In contrast to most previous studies, this work focusses on analytical derivations. To this end, the Cramér-Rao bound of the best attainable state estimation covariance is derived from a model of sensor measurement errors. This state- and time-dependent covariance is then propagated to an AEB decision making logic that is based on a criticality measure. Additional inherent prediction uncertainty in this risk assessment is taken into account. The effectiveness of the AEB subject to uncertainties is compared to the deterministic baseline case in terms of the brake activation time and the collision energy reduction.

Statistical Modelling of Object Detection in Stereo Vision-Based Driver Assistance

In this work, a statistical analysis of object detection for stereo vision-based driver assistance systems is presented. Analytic modelling has not been attempted previously due to the complexity of dense disparity maps and state-of-the-art algorithms. To approach this problem, a simplified algorithm for object detection in stereo images which allows studying error propagation is considered. In order to model the input densities, vehicle contours are approximated by Gaussian Mixture Models and distance dependent measurement noise is taken into account. Theoretical results are verified with Monte Carlo methods and real-world image sequences. Using the proposed model, a prediction on the uncertainty in object location and optimal threshold selection can be obtained.

Performance bounds on change detection with application to manoeuvre recognition for advanced driver assistance systems

Recognising the intended manoeuvres of other traffic participants is a crucial task for situation interpretation in driver assistance and autonomous driving. While many works propose algorithms for (computationally feasible) inference, much less attention is paid to finding analytic upper performance bounds for these problems. This work studies the statistical properties of the optimal detector in a binary change detection problem, i.e. the Generalised Likelihood Ratio test. With analytic models of the best attainable receiver operating characteristic, the influence of system design parameters can be investigated without the need for empirical evaluation. Moreover, these bounds can be used to derive objective performance metrics.

Development Strategy for Collision Warning Systems in the Low Price Segment

In the field of safety oriented driver assistance systems, there is the tendency to reduce accidents and their consequences by autonomous interventions in the driving process which are getting stronger and stronger. But this increase of the “effectiveness” also increases the danger emanating from such a system in case of a malfunction. This increases the effort of development and safeguarding and with it also severely the costs of such functions. It is possible however to develop systems with comparable effectiveness without using strong autonomous interventions. For this, Bosch combined very early collision warning with driver assisting functions. This provides the driver with sufficient time to assess the situation and assists in initiating measures for accident prevention in the best possible way. The technical challenge is the realization of an early and at the same time reliable criticality assessment of the traffic situation.

Entwicklungsstrategie für Kollisionswarnsysteme im Niedrigpreis-Segment

Im Bereich der sicherheitsorientierten Fahrerassistenzsysteme besteht ein Trend, durch immer starkere autonome Eingriffe in das Fahrgeschehen Unfalle und deren Folgen zu vermindern. Diese Erhohung des „Wirkgrads“ vergrosert jedoch auch die Gefahr, die von einem solchen System im Fall einer Fehlfunktion ausgeht. Dies erhoht den Entwicklungs- und Absicherungsaufwand und damit auch sehr stark die Kosten solcher Funktionen. Unter Verzicht starker autonomer Eingriffe lassen sich jedoch Systeme mit vergleichbarem Wirkgrad entwickeln. Dazu hat Bosch eine sehr fruhe Kollisionswarnung mit fahrerunterstutzenden Funktionen kombiniert. Der Fahrer erhalt damit ausreichend Zeit zur Bewertung der Situation und wird bei der Einleitung von unfallvermeidenden Masnahmen optimal unterstutzt. Die technische Herausforderung besteht in der Realisierung einer fruhen und gleichzeitig zuverlassigen Kritikalitatsbewertung der Verkehrssituation.