gregkuhlmann

Mohan Sridharan, Gregory Kuhlmann, and Peter Stone
In IEEE International Conference on Robotics and Automation (ICRA-05), September 2004.

Abstract: Mobile robot localization, the ability of a robot to determine its position and orientation in a global frame of reference, continues to be a major research focus in robotics. In most past cases, such localization has been studied on wheeled robots with range-finding sensors such as sonar or lasers. In this paper, we consider the more challenging scenario of a legged robot localizing with limited-field-of-view vision as the primary sensory input. We begin with a baseline implementation adapted from the literature that provides a reasonable level of competence, but that exhibits some weaknesses in real-world tests. We propose a series of practical enhancements designed to improve the robot's sensory and actuator models that enable our robots to achieve a 50% improvement in localization accuracy over the baseline implementation, and even more dramatic improvements when the robot is subjected to large unexpected movements. These enhancements are each individually straightforward, and they do not change the basic particle filtering approach. But together they provide a practical guide for avoiding potential pitfalls when implementing it on vision-based and/or legged robots. Our complete localization system is fully implemented on the Sony ERS-7 robot platform. We present extensive empirical results, both in simulation and on the physical robots, isolating the impacts of our contributions.

@InCollection(ICRA2005-localization,
  author="Mohan Sridharan and Gregory Kuhlmann and Peter Stone",
  title="Practical Vision-Based Monte Carlo Localization on a Legged Robot",
  booktitle="IEEE International Conference on Robotics and Automation (ICRA-05)",
  month="September",
  year="2004"
)