Felix Klanner

Driver intent inference at urban intersections using the intelligent driver model

Predicting turn and stop maneuvers of potentially errant drivers is a basic requirement for advanced driver assistance systems for urban intersections. Previous work has shown that an early estimate of the drivers intent can be inferred by evaluating the vehicles speed during the intersection approach. In the presence of a preceding vehicle, however, the velocity profile might be dictated by car-following behaviour rather than by the need to slow down before doing a left or right turn. To infer the drivers intent under such circumstances, a simple, real-time capable approach using an explicit model to represent both car-following and turning behaviour is proposed. Models for typical turning behavior are extracted from real world data. Preliminary results based on a Bayes net classification are presented.

Velocity-Based Driver Intent Inference at Urban Intersections in the Presence of Preceding Vehicles

Predicting turn and stop maneuvers of potentially errant drivers is a basic requirement for advanced driver assistance systems for urban intersections. Previous work has shown that an early estimate of the drivers intent can be inferred by evaluating the vehicles speed during the intersection approach. In the presence of a preceding vehicle, however, the velocity profile might be dictated by car-following behavior rather than by the need to slow down before doing a left or right turn. To infer the drivers intent under such circumstances, a simple, real-time capable approach using a parametric model to represent both car-following and turning behavior is proposed. The performance of two alternative parameterizations based on observations at an individual intersection and a generic curvature-based model is evaluated in combination with two different Bayes net classification algorithms. In addition, the driver model is shown to be capable of predicting the future trajectory of the vehicle.

Car2X-based perception in a high-level fusion architecture for cooperative perception systems

In cooperative perception systems, different vehicles share object data obtained by their local environment perception sensors, like radar or lidar, via wireless communication. In this paper, this so-called Car2X-based perception is modeled as a virtual sensor in order to integrate it into a highlevel sensor data fusion architecture. The spatial and temporal alignment of incoming data is a major issue in cooperative perception systems. Temporal alignment is done by predicting the received object data with a model-based approach. In this context, the CTRA (constant turn rate and acceleration) motion model is used for a three-dimensional prediction of the communication partners motion. Concerning the spatial alignment, two approaches to transform the received data, including the uncertainties, into the receiving vehicles local coordinate frame are compared. The approach using an unscented transformation is shown to be superior to the approach by linearizing the transformation function. Experimental results prove the accuracy and consistency of the virtual sensors output.

Active safety for vulnerable road users based on smartphone position data

Smartphones have long become an omnipresent part of our life. Equipped with both a broadband internet connection and advanced GPS onboard sensors, the idea is to use them as mobile sensors for active safety systems that aim at protecting vulnerable road users such as pedestrians or cyclists. This paper gives a comprehensive analysis of todays smartphones GPS accuracy on an inner-city bicycle track. In addition, the transmission latencies of a prototypical bicycle warning system are evaluated. The results show that while the lateral deviations are still too high to allow for lane-level localization, the longitudinal accuracy as well as the transmission latencies are good enough for many active safety applications already.

Analysis of V2X communication parameters for the development of a fusion architecture for cooperative perception systems

In cooperative perception systems, different vehicles share object data obtained by their local environment perception sensors, like radar or lidar, via wireless communication. In this paper, a fusion architecture for a cooperative perception system and a concept for a parametrizable offline simulation are proposed. In order to effectively develop and simulate such systems, knowledge about the main parameters of the employed wireless communication solution is crucial. For this reason, an experimental analysis of parameters like transmission latencies and transmission range of a communication solution based on IEEE 802.11p is presented.

Inter-vehicle object association for cooperative perception systems

In cooperative perception systems, different vehicles share object data obtained by their local environment perception sensors, like radar or lidar, via wireless communication. Inaccurate self-localizations of the vehicles complicate association of locally perceived objects and objects detected and transmitted by other vehicles. In this paper, a method for inter-vehicle object association is presented. Position and orientation offsets between object lists from different vehicles are estimated by applying point matching algorithms. Different algorithms are analyzed in simulations concerning their robustness and performance. Results with a first implementation of the so-called Auction-ICP algorithm in a real test vehicle validate the simulation results.

Generic driver intent inference based on parametric models

Reasoning about the driver intent is fundamental both to advanced driver assistance systems as well as to highly automated driving. In contrast to the vast majority of preceding work, we investigate an architecture that can deal with arbitrary combinations of subsequent maneuvers as well as a varying set of available features. Detailed parametric models are given for the indicator, velocity and gaze direction features, all of which are parametrized from the results of extensive user studies. Evaluation is carried out for continuous right-turn prediction on a separate data set. Assuming conditional independence between the individual feature likelihoods, we investigate the contribution of each feature to the overall classification result separately. In particular, the approach is shown to work well even when faced with implausible observations of the indicator feature.

Crowdsourced intersection parameters: A generic approach for extraction and confidence estimation

Digital maps within cars are not only the basis for navigation but also for advanced driver assistance systems. Therefore more and more up-to-date details about the environment of the vehicle are required which means that they have to be enriched with further attributes such as detailed representations of intersections. In the future we will be able to extract details of the environment out of the sensory data of connected cars. We present a generic approach for extracting multiple intersection parameters with the same method by analyzing logged data from a test fleet. Based on that a method for a feature based estimation of the confidence is introduced. The proposed approaches are applied in a completely automated process to estimate stop line positions and traffic flows at intersections with traffic lights. Altogether 203.701 traces of the test fleet were used for developing and testing. The performance of the method and the confidence estimation were analyzed using a ground truth, consisting of 108 stop line positions, which was derived from satellite images. The results show that the approach is fast and predictions with an absolute accuracy of 3.5m can be achieved. Hence the method is able to deliver valuable inputs for driver assistance systems.

Vehicle Mass Estimation Based on Vehicle Vertical Dynamics Using a Multi-Model Filter

Vehicle mass estimation is an important task to compute the input parametrization for various advanced driver assistance systems. Further, detecting when a trailer is present or even the mass distribution between vehicle and trailer is of interest for various systems. Thus, we discuss the influence between vehicle and trailer of different approaches and finally yield the mass distribution over the vehicle-trailer-combination by linking mass estimates from longitudinal and vertical dynamics. This work investigates a multi-model approach for vehicle mass estimation based on common sensor signals for vertical dynamics as they are available in modern vehicle suspension systems with no need of a calibrated reference. For tracking the suspension dynamics, we partly use a Kalman filter with a linear dynamic system matrix. However, it is not feasible to gain consistent filter behavior for all possible mass hypotheses. Thus, we apply modifications to the common multi-model approach and define an evaluation function for the model probabilities to overcome the consistency issue and speed up computation time.

Route and Stopping Intent Prediction at Intersections From Car Fleet Data

In this paper, an approach is presented to predict the route and stopping intent of human-driven vehicles at urban intersections using a selection of distinctive features observed on the vehicle state (position, heading, acceleration, velocity). For potential future advanced driver assistance systems, this can facilitate the situation analysis and risk assessment at road intersections, helping to improve the protection of vulnerable road users. After extracting recorded driving data for nine intersections (featuring over 50 000 crossings) from a database, they are assigned to possible routes and transformed from a time-based representation to a distance-based one. Using random decision forests, the route intent can be predicted with a mean unweighted average recall (UAR) of 0.76 at 30 m before the relevant intersection center, the stopping intent prediction scores a mean UAR of 0.78.