The International Journal of Robotics Research · 1999 · 359 citations · 21 references
EngineeringComputer ScienceRobotics EngineeringMobile RoboticsRobot ControlRoboticsAutonomous ExplorationAutonomous NavigationMappingRobot LocalizationRobot NavigationUnderwater LocalizationDead ReckoningExtended Kalman FilterFisher InformationFisher Information AnalysisMobile Robot NavigationRobot MappingRobotic LocalizationUnderwater Vehicle Navigation
Building a map of an unknown environment while simultaneously navigating it is a central challenge in mobile robotics, often addressed by stochastic feature‑based mapping that extends the extended Kalman filter. This paper seeks to perform concurrent mapping and localization adaptively using sonar. The authors implement stochastic mapping with delayed nearest‑neighbor data association, compute a Fisher‑information‑based adaptive sensing metric from predicted sensor readings and dead‑reckoning errors, and select the action minimizing this metric, demonstrating the approach in simulations, in‑air sonar, and underwater sonar experiments. Adaptive control of motion and scanning leads the vehicle to selectively explore different objects, and the algorithm outperforms straight‑line and random motion in performance.
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. This paper addresses the problem of how to perform concurrent mapping and localization (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 implementation of stochastic mapping that uses a delayed nearest neighbor data association strategy to initialize new features into the map, match measurements to map features, and delete out-of-date features. The authors introduce a metric for adaptive sensing that is defined in terms of Fisher information and represents the sum of the areas of the error ellipses of the vehicle and feature estimates in the map. Predicted sensor readings and expected 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 simulations, in-air sonar experiments, and underwater sonar experiments. Results are shown for (1) adaptive control of motion and (2) adaptive control of motion and scanning. The vehicle tends to explore selectively different objects in the environment. The performance of this adaptive algorithm is shown to be superior to straight-line motion and random motion.
21
<i>Tracking and Data Association</i>
Yaakov Bar‐Shalom, Thomas Fortmann, Peter G. Cable · The Journal of the Acoustical Society of America · 1990 · 3.1K citations · Full text
Computing, Mathematics, Physics +14
Artificial Intelligence: A Modern Approach
Stuart Russell, Peter Norvig · RPK (Politechniki Krakowskiej) · 1995 · 1.8K citations · Full text