Robotics and Sensor-Based Localization

Robots and autonomous vehicles can only act reliably in the world if they know where they are and what surrounds them, which is precisely what sensor-based localization research addresses. Simultaneous Localization and Mapping, or SLAM, tackles both problems at once: a system builds a geometric model of its environment — often as a dense point cloud — while continuously estimating its own position within that model, using data from cameras, lidar, or inertial sensors. In aerospace contexts, where GPS is unreliable or absent entirely, techniques like visual odometry and RGB-D mapping become critical for spacecraft, drones, and planetary rovers operating in unknown terrain. Active research is pushing toward more robust performance under challenging lighting and motion conditions, tighter real-time constraints on embedded hardware, and the fusion of multiple sensor modalities to reduce the drift that accumulates over long trajectories.

Works

91,472

Total citations

1,379,641

Keywords

SLAM3D MappingVisual OdometryRoboticsLocalizationPoint Cloud