Position and Orientation Errors in Mobile Robot Absolute Self-Localization Using an Improved Version of the Generalized Geometric Triangulation Algorithm

Abstract

Triangulation with active beacons is widely used in the absolute localization of mobile robots. The original Generalized Geometric Triangulation algorithm suffers only from the restrictions that are common to all algorithms that perform self-localization through triangulation. But it is unable to compute position and orientation when the robot is over the segment of the line that goes by beacons 1 and 2 whose origin is beacon 1 and does not contain beacon 2. An improved version of the algorithm allows self-localization even when the robot is over that line segment. Simulations results suggest that a robot is able to localize itself, with small position and orientation errors, over a wide region of the plane, if measurement uncertainty is small enough.

BeAMS: A Beacon-Based Angle Measurement Sensor for Mobile Robot Positioning

“…It is a well known fact that a positioning process based on angles, regardless of its implementation, depends on the relative configuration of beacons and the robot [10], [14], [15]. So, we believe that an angle measurement system should not be evaluated through a positioning algorithm, unless a common procedure is described and used by everyone.…”

“…Some fundamental papers such as [5], [6], [8], [38] discuss robot positioning. In particular, Betke and Gurvits [5] and Esteves et al [15] highlight that sensory feedback is essential in order to position the robot in its environment. Some surveys (see [7], [12], [18], [28]) discuss several techniques used for positioning: odometry, inertial navigation, magnetic compasses, active beacons, natural landmark navigation, map-based positioning, and vision-based positioning.…”

“…Because of its robustness, accuracy, and flexibility, triangulation with active beacons is widely used for robots [38]. Another advantage of triangulation versus trilateration is that the robot can compute its orientation (heading) in addition to its location [15], [17], [34], which can be as important as the robot position for most applications.…”

“…Triangulation methods using three angle measurements can be found in [10], [14], [15], [17], [27], [34], and methods using more than three angle measurements are described in [5], [38]. More than three angles can be used to increase precision in some pathological Bi are the beacons.…”

Methodical Approach to Active Node Layout Generation for Robot Navigation

“…There are a variety methods by which this can be done, such as some form of triangulation or the overlapping of sensing ranges [1][2][3][4]. This concept is of course similar to approaches often used in vision-based navigation where, using SIFT [5] for example, landmarks are identified in a room and tracked.…”