Session Overview Planning

ys when planning meant searching for a sequence of abstract actions that satisfied some symbolic predicate. Robots can now learn their own representations through statistical inference procedures, they can now reason using different representations and they can reason in worlds where action can have stochastic outcomes. However, despite the successes of robots that use machine learning and statistical inference in such different areas as mapping, speech recognition, computer vision, etc., there remain open questions to be addressed before we will see ubiquitous, useful, mobile robots, and some of the most interesting problems are in the planning domain. Consider a mobile robot deployed in some populated environment such as the home. A human operator typically drives the robot around in order to collect sensor data. This data is then used to build a “good” map that is largely static. The robot planner then computes good paths through this map, assuming that the map is correct and...

• Type: Conference
• Year: 2005
• Where: ISRR
• Authors: Nicholas Roy, Roland Siegwart
• Keywords: ISRR 2005, Mobile Robots, Robotics, Robots, Statistical Inference