Information-Fusion Methods Based Simultaneous Localization and Mapping for Robot Adapting to Search and Rescue Postdisaster Environments

Abstract

The first application of utilizing unique information-fusion SLAM (IF-SLAM) methods is developed for mobile robots performing simultaneous localization and mapping (SLAM) adapting to search and rescue (SAR) environments in this paper. Several fusion approaches, parallel measurements filtering, exploration trajectories fusing, and combination sensors' measurements and mobile robots' trajectories, are proposed. The novel integration particle filter (IPF) and optimal improved EKF (IEKF) algorithms are derived for information-fusion systems to perform SLAM task in SAR scenarios. The information-fusion architecture consists of multirobots and multisensors (MAM); multiple robots mount on-board laser range finder (LRF) sensors, localization sonars, gyro odometry, Kinect-sensor, RGB-D camera, and other proprioceptive sensors. This information-fusion SLAM (IF-SLAM) is compared with conventional methods, which indicates that fusion trajectory is more consistent with estimated trajectories and real observation trajectories. The simulations and experiments of SLAM process are conducted in both cluttered indoor environment and outdoor collapsed unstructured scenario, and experimental results validate the effectiveness of the proposed informationfusion methods in improving SLAM performances adapting to SAR scenarios.

Collaborative Multi-Robot Search and Rescue: Planning, Coordination, Perception, and Active Vision

“…Simultaneous localization and mapping (SLAM) aims at constructing or updating a map of the environment of an agent, while simultaneously keeping track of the agent's position. In SLAM, RGB-D data is used to build a dense 3D map and the data fusion technique applied in a single-agent SLAM is typically extended Kalman Filter (EKF) [215]. Fusing RGB and thermal image data can be needed, for example, in man overboard situations [216].…”

Collaborative Multi-Robot Systems for Search and Rescue: Coordination and Perception

“…Simultaneous localization and mapping (SLAM) aims at constructing or updating a map of the environment of an agent, while simultaneously keeping track of the agent's position. In SLAM, RGB-D data is used to build a dense 3D map and the data fusion technique applied in a single-agent SLAM is typically extended Kalman Filter (EKF) [214]. Fusing RGB and thermal image data can be needed, for example, in man overboard situations [215].…”