Volker Willert

A Probabilistic Model for Binaural Sound Localization

This paper proposes a biologically inspired and technically implemented sound localization system to robustly estimate the position of a sound source in the frontal azimuthal half-plane. For localization, binaural cues are extracted using cochleagrams generated by a cochlear model that serve as input to the system. The basic idea of the model is to separately measure interaural time differences and interaural level differences for a number of frequencies and process these measurements as a whole. This leads to two-dimensional frequency versus time-delay representations of binaural cues, so-called activity maps. A probabilistic evaluation is presented to estimate the position of a sound source over time based on these activity maps. Learned reference maps for different azimuthal positions are integrated into the computation to gain time-dependent discrete conditional probabilities. At every timestep these probabilities are combined over frequencies and binaural cues to estimate the sound source position. In addition, they are propagated over time to improve position estimation. This leads to a system that is able to localize audible signals, for example human speech signals, even in reverberating environments

Estimating Object Proper Motion Using Optical Flow, Kinematics, and Depth Information

For the interaction of a mobile robot with a dynamic environment, the estimation of object motion is desired while the robot is walking and/or turning its head. In this paper, we describe a system which manages this task by combining depth from a stereo camera and computation of the camera movement from robot kinematics in order to stabilize the camera images. Moving objects are detected by applying optical flow to the stabilized images followed by a filtering method, which incorporates both prior knowledge about the accuracy of the measurement and the uncertainties of the measurement process itself. The efficiency of this system is demonstrated in a dynamic real-world scenario with a walking humanoid robot.

DisCoverage: A new paradigm for multi-robot exploration

The main aspect in multi-robot exploration is the efficient coordination of a group of robots. Inspired by previous results on the coverage problem, we propose a novel, frontier-based approach for multi-robot exploration. This approach merges the step of choosing appropriate target points with the step of planning a collision-free path. This is achieved by optimizing an objective function consisting of distance and orientation costs as well as an estimated information gain. The optimization yields motion control laws directly solving the exploration task. Using a Voronoi partition of the environment ensures, that each robot autonomously creates and optimizes the objective function to obtain a collision-free path in a distributed fashion. Simulations demonstrate the effectiveness of our approach.

PRORETA 3: An Integrated Approach to Collision Avoidance and Vehicle Automation

Zusammenfassung The article describes first results of the research project PRORETA 3 that aims at the development of an integral driver assistance system for collision avoidance and automated vehicle guidance based on a modular system architecture. For this purpose, relevant information is extracted from a dense environment model and fed into a potential field-based trajectory planner that calculates reference signals for underlying vehicle controllers. In addition, the driver is supported by a human-machine interface. Abstract Der Beitrag beschreibt erste Ergebnisse des Forschungsprojektes PRORETA 3, das die Entwicklung eines integralen Fahrerassistenzsystems zur Kollisionsvermeidung und automatisierten Fahrzeugführung auf Basis einer modularen Systemarchitektur anstrebt. Hierzu werden relevante Informationen aus einem dichten Umfeldmodell extrahiert und in einem potentialfeldbasierten Trajektorienplaner verarbeitet, der Führungsgrößen für unterlagerte Fahrzeugregler generiert. Zusätzlich unterstützt eine Mensch-Maschine-Schnittstelle den Fahrer zielgerichtet bei der Fahrzeugführung.

Non-Gaussian velocity distributions integrated over space, time, and scales

Velocity distributions are an enhanced representation of image velocity containing more velocity information than velocity vectors. In particular, non-Gaussian velocity distributions allow for the representation of ambiguous motion information caused by the aperture problem or multiple motions at motion boundaries. To resolve motion ambiguities, discrete non-Gaussian velocity distributions are suggested, which are integrated over space, time, and scales using a joint Bayesian prediction and refinement approach. This leads to a hierarchical velocity-distribution representation from which robust velocity estimates for both slow and high speeds as well as statistical confidence measures rating the velocity estimates can be computed.

Bayesian, maneuver-based, long-term trajectory prediction and criticality assessment for driver assistance systems

We propose a Bayesian trajectory prediction and criticality assessment system that allows to reason about imminent collisions of a vehicle several seconds in advance. We first infer a distribution of high-level, abstract driving maneuvers such as lane changes, turns, road followings, etc. of all vehicles within the driving scene by modeling the domain in a Bayesian network with both causal and diagnostic evidences. This is followed by maneuver-based, long-term trajectory predictions, which themselves contain random components due to the immanent uncertainty of how drivers execute specific maneuvers. Taking all uncertain predictions of all maneuvers of every vehicle into account, the probability of the ego vehicle colliding at least once within a time span is evaluated via Monte-Carlo simulations and given as a function of the prediction horizon. This serves as the basis for calculating a novel criticality measure, the Time-To-Critical-Collision-Probability (TTCCP) - a generalization of the common Time-To-Collision (TTC) in arbitrary, uncertain, multi-object driving environments and valid for longer prediction horizons. The system is applicable from highly-structured to completely non-structured environments and additionally allows the prediction of vehicles not behaving according to a specific maneuver class.

Robust free space detection in occupancy grid maps by methods of image analysis and dynamic B-spline contour tracking

We propose a new method for free space detection and description for Advanced Driver Assistance Systems (ADAS) and autonomous vehicles. The detection is based on successive morphological image processing steps that are applied to an occupancy grid map-based environment representation acquired by an automotive radar sensor. The boundary of the found free space segment is traced and serves as a virtual measurement for a time-variant Kalman Filter in order to estimate and track the control points of a two-dimensional B-spline closed free space contour over time. In contrast to existing free space detection methods, the proposed solution incorporates knowledge about the vehicles dimensions and does not exclude free space that is not directly in the line of sight, but mapped beforehand, as well as free space behind obstacles. Furthermore, the algorithm shows advantages in terms of an intuitive control over spatial and temporal smoothness of the solution as well as an inherent robustness due to model-based filtering. Moreover, the control points of the B-spline curve are proposed as a new low-dimensional representation of drivable free space of arbitrary shape. The effectiveness of the algorithm is demonstrated in real traffic scenarios.

Combining Behavior and Situation Information for Reliably Estimating Multiple Intentions

Intersections are the most accident-prone spots in the road network. In order to assist the driver in complex urban intersection situations, an ADAS will be required not only to recognize current but also to anticipate future maneuvers of the involved road users. Current approaches for intention estimation focus mainly on discerning only two intentions based on a vehicles behavior. We argue that for distinguishing between more than two intentions not just a vehicles kinematic behavior but also its driving situation needs to be taken into account. In our system we estimate four different intentions by modeling and recognizing driving situations in a Bayesian Network and using the behavior as additional evidence. For the behavior based estimation we present a newly engineered feature, the Anticipated Velocity at Stop line, that turned out to be a very strong indicator for the intention. Our system is evaluated on a real-world data set comprising approaches to seven different intersections on which we can show that our approach is able to estimate a drivers intention with a high accuracy.

How to distinguish inliers from outliers in visual odometry for high-speed automotive applications

In this paper, we present an outlier removal scheme for stereo-based visual odometry which is especially suited for improving high-speed pose change estimations in large-scale depth environments. First we investigate the variance of the reprojection error on the 3D position of a feature given a fixed error in pose change to conclude that a detection of outliers based on a fixed threshold on the reprojection error is inappropriate. Then we propose an optical flow dependent feature-adaptive scaling of the reprojection error to reach almost invariance to the 3D position of each feature. This feature-adaptive scaling is derived from an approximation showing the relation between longitudinal pose change of the camera, absolute value of the optical flow, and distance of the feature. Using this scaling, we develop an iterative alternating scheme to guide the separation of inliers from outliers. It optimizes the tradeoff between finding a good criterion to remove outliers based on a given pose change and improving the pose change hypothesis based on the current set of inliers. Including the new outlier removal scheme into a pure two-frame stereo-based visual odometry pipeline without applying bundle adjustment or SLAM-filtering we are currently ranked amongst the top camera-based algorithms and furthermore outperform camera and laser scanner methods in Kitti benchmarks high-speed scenarios.

Non-negative sparse coding for motion extraction

Visual motion is a rich source of information that is directly coupled to the underlying shape of a moving object. One way to describe motion is to use optical flow fields. Due to the aperture problem, dense optical flow estimation is an ill-constraint problem, while sparse optical flow estimation looses the shape information of moving objects. Current estimation algorithms based on regularization or segmentation fail at surface deformations or when the relevant motion is less dominant then its sourrounding movements. Both is e.g. true for face movements, where small movement patterns, so called action units, need to be preserved for further image analysis. We present a novel approach to capture the characteristics of local motion patterns that is based on the brightness constancy equation of optical flow estimation in combination with feature extraction using translation invariant non-negative sparse coding. Our approach simultaneously learns basic motion patterns and estimates the flow field without requiring pretrained motion patterns from ground truth optical flow data. We show on a face expression dataset how this method can preserve weak movements even in the presence of large head movements.