Skip to Main Content
Over the past decade, challenging applications for autonomous robots have been identified, in the areas of servicing crowded, built-up areas, mining, search and rescue operations, underwater exploration and airborne surveillance. Autonomous navigation arguably remains the key enabling issue behind any realistic commercial success in these areas. Consequently, autonomous robotic research has focused on large scale and long term navigation algorithms, sensing technologies, robust sensor data interpretation and map building. The recent breakthroughs which contribute to the success of outdoor field robotics, and remaining fundamental research issues involved, will be the theme of this presentation. The most successful robot navigation algorithms to-date, have been derived from a probabilistic perspective, which takes into account vehicle motion and terrain uncertainty and sensor noise. Over the past decade, an explosion of interest in the estimation of an autonomous robot's location state, and that of its surroundings, known as simultaneous localisation and map building (SLAM), is evident. New algorithms which represent uncertain information based on particle filters and Gaussian mixture models, as well as the more classical Kalman filter based techniques, are advancing the progress of a robot's long term navigation abilities. This has been significantly aided by recently affordable sensor technologies, including GPS and inertial measurement units (IMUs) as well as fast and reliable laser range finders.