Adaptive mobile robot navigation and mapping

The task of building a map of an unknown environment and concurrently using that map to navigate is a central problem in mobile robotics research. Thi s paper addresses the problem of how to perform concurrent mapping and loca lization (CML) adaptively using sonar Stochastic mapping is a feature-based approach to CML that generalizes the extended Kalman filter to incorporate vehicle localization and environmental mapping. The authors describe an im plementation of stochastic mapping that uses a delayed nearest neighbor dat a association strategy to initialize new features into the map, match measu rements to map features, and delete out-of-date features. The authors intro duce a metric for adaptive sensing that is defined in terms of Fisher infor mation and represents the sum of the areas of the error ellipses of the veh icle and feature estimates in the map Predicted sensor readings and expecte d dead-reckoning errors are used to estimate the metric for each potential action of the robot, and the action that yields the lowest cost (i.e., the maximum information) is selected. This technique is demonstrated via simula tions, in-air sonar experiments, and underwater sonar experiments. Results are shown for (I) adaptive control of motion and(2) adaptive control of mot ion and scanning. The vehicle tends to explore selectively different object s in the environment. The performance of this adaptive algorithm is shown t o be superior to straight-line motion and random motion.