Simultaneous Localisation and Mapping

The problem of simultaneous localisation and mapping (SLAM) for an autonomous mobile robot is stated as follows: starting from a initial position, a mobile robot travels through a sequence of positions and obtains a set of sensor measurements at each position. The goal is for the mobile robot to process the sensor data to produce an estimate of its position while concurrently building a map of the environment. During the last decade SLAM has been a significant research problem, that is now considered solved in simple 2D scenarios using laser scanners. The main open challenges include efficient mapping of large environments, modelling complex and dynamic environments, multi-vehicle SLAM and full 3D SLAM. Most of this challenges will require scalable representations, robust data association algorithms, consistent estimation techniques, and different sensor modalities. In particular, solving SLAM with monocular or stereo vision is a crucial research goal for addressing many real life applications. The objectives of this workshop are to present recent advances on these key issues, discuss and compare different approaches and assess the current state-ofthe-art in SLAM. The workshop will also cover a wide variety of classical and emerging SLAM applications such as field and service robots, off-road driving, mining, underwater, airborne, humanoids and augmented reality. ICRA'2005 Workshop on SLAM 1