Contents Acknowledgements .............................................................................. i Abstract............................................................................................... iii Contents ................................................................................................v 1

Abstract — Situational awareness is crucial for autonomous driving in urban environments. This paper describes moving vehicle tracking module that we developed for our autonomous driving robot Junior. The robot won second place in the Urban Grand Challenge, an autonomous driving race organized by the U.S. Government in 2007. The tracking module provides reliable tracking of moving vehicles from a high-speed moving platform using laser range finders. Our approach models both dynamic and geometric properties of the tracked vehicles and estimates them using a single Bayes filter per vehicle. We also show how to build efficient 2D representations out of 3D range data and how to detect poorly visible black vehicles. Experimental validation includes the most challenging conditions presented at the UGC as well as other urban settings. I.

Most approaches for vision systems use greyscale or color images. In many applications, such as driver assistance or presence detection systems, the geometry of the scene is more relevant than the reected brightness information and therefore range sensors are of increasing interest.

The weakest link in many mobile robots is perception. In order to build robots that are reliable and dependable and safe, we need to build robots that can see. Perception is becoming a solved problem for certain constrained environments. But for robots working outdoors, and at high speeds, and in close proximity to people, perception is still incomplete. Our robots need to see objects; to detect motion; and to detect which of those objects are people. In the current state of the art, this requires multiple sensors and multiple means of interpretation. This paper illustrates those principles in the context of the CMU Navlab Group's work on vehicle safety for busses and passenger cars. 1

Classical vehicle tracking approaches for highway scenarios use a Kalman--filter with a single dynamic model optimised to a single driving manoeuvre. In contrast, the Interacting Multiple Model (IMM) filter allows for several parallel models which are combined to a weighted estimate. Choosing models for different driving modes such as constant speed, acceleration and strong acceleration changes permits the object state estimation to be optimised for highly dynamic driving manoeuvres. The paper describes the analysis of Stop&Go situations and the systematic parametrisation of the IMM method based on these statistics. The evaluation of the IMM approach is presented based on real sensor measurements of two laserscanners, a radar and a video image processing unit. The performance of the lateral estimation of the IMM is shown based on simulations.

Abstract — Situational awareness is crucial for autonomous driving in urban environments. We present the moving vehicle tracking module we developed for our autonomous driving robot Junior. The robot won second place in the Urban Grand Challenge, an autonomous driving race organized by the U.S. Government in 2007. The module provides reliable detection and tracking of moving vehicles from a high-speed moving platform using laser range finders. Our approach models both dynamic and geometric properties of the tracked vehicles and estimates them using a single Bayes filter per vehicle. We show how to build consistent and efficient 2D representations out of 3D range data and how to detect poorly visible black vehicles. Experimental validation includes the most challenging conditions presented at the Urban Grand Challenge as well as other urban settings. I.

One of the challenges in the automation of vehicles takes place in the car park. Automatic vehicles must be able to localize itself accurately and build a map of the car park in real time. This capability will be able to extend the range of possible applications including the tracking of car park lots availability, automatic navigation and automatic parking to name a few. In this project, we took on the task of localizing an automatic vehicle and building a map of the car park in real time. This project takes place within the car park of INRIA Rhone-Alpes on the CyCab vehicle with a Sick laser range scanner. A key feature is that it works only with laser scanners to retrieve the position and orientations of vehicles in the car park. With the detected vehicles as landmarks, CyCab performs a localization of itself and build a map of the car park at the same time. This problem is commonly known in the literature as Simultaneous Localization and Mapping(SLAM). The detection of vehicles is based on the work of a previous DEA student (Lorieux) [11] and extensions were proposed to increase the reliability of vehicle detection. The SLAM algorithm chosen is the FastSLAM algorithm. The FastSLAM algorithm is adapted within this context. However, FastSLAM only gives a set of hypotheses. Hence, a map construction method is proposed to merge the different hypotheses together to form one single final map. Experiments with real data were conducted and it is able to perform the tasks required. The experiments also revealed weaknesses in the system and possible approaches and directions for further research is suggested. Contents

in my opinion, it