Joakim Lin Sorstedt

Extended Object Tracking using a Radar Resolution Model

The work presented here concerns the problem of vehicle tracking when multiple radar reflection centers could be resolved on each vehicle. For this extended target tracking problem we propose a radar sensor model, capable of describing such measurements, incorporating sensor resolution. Furthermore, we introduce approximations to handle the inherently complex data association problem. The evaluation in terms of describing measured data and resulting tracking performance shows that the model effectively exploits the information in multiple vehicle detections.

A New Vehicle Motion Model for Improved Predictions and Situation Assessment

Reliable and accurate vehicle motion models are of vital importance for automotive active safety systems for a number of reasons. First of all, these models are necessary in tracking algorithms that provide the safety system with information. Second, the motion model is often used by the safety application to make long-term predictions about the future traffic situation. These predictions are then part of the basic data used by the system to determine if, when, and how to intervene. In this paper, we suggest a framework for designing accurate vehicle motion models. The resulting models differ from conventional models in that the expected control input from the driver is included. By also providing a methodology for a formal treatment of the uncertainties, a model structure well suited, e.g., in a tracking algorithm, is obtained. To utilize the framework in an application will require careful design and validation of submodels to calculate the expected driver control input. We illustrate the potential of the framework by examining the performance for a specific model example using real measurements. The properties are compared with those of a constant acceleration model. Evaluations indicate that the proposed model yields better predictions and that it has an ability to estimate the prediction uncertainties.

Sensor data fusion for multiple configurations

This paper presents a fusion architecture designed for vehicle manufacturers that use multiple sensor systems to realize several active safety applications, ranging from standard systems to autonomous systems. Advantages and disadvantages of design choices are discussed and the methods we have chosen to implement in two demonstrator vehicles are used as examples. The main contributions are the separation of the architecture into three categories, the state vector parametrization, and the use of multiple output-lists. Together, these choices make it possible to re-use verified core functionality while filters can be independently tuned to meet application-specific requirements.

Ego lane estimation using vehicle observations and map information

An ego vehicle localization algorithm must be able to estimate where the vehicle is on the road. This is typically performed with a positioning filter that operates in global coordinates. Herein, we take a different approach, by splitting the localization problem into two parts: in-lane localization and ego lane estimation. The paper addresses the latter problem. For this, we have developed theory and algorithms which, based on information about the positions of surrounding vehicles, give the probability of being in each of the current number of lanes. The object positions are provided by one or several low-cost on-board perception sensors. The derived Bayesian filter is evaluated on real data from a prototype self-driving car. Preliminary results show that when other vehicles are present, the proposed method is able to estimate the lane of travel with high probability.