Share Email Print
cover

Proceedings Paper

Simultaneous localization and mapping with consideration of robot system dynamics
Author(s): R. Jaai; N. Chopra; B. Balachandran; H. Karki
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

In the simultaneous localization and mapping (SLAM) problem, it is required for a robotic system to acquire the map of its environment while simultaneously localizing itself relative to this evolving map. In order to solve the SLAM problem, given observations of the environment and control inputs, the joint posterior probability of the robot pose and the map are estimated by using recursive filters such as the extended Kalman filter (EKF) and the particle filter. The implementation of these filters requires a motion model to describe the evolution of the robot pose with control inputs, and additionally, an observation model to describe the relations between the robot pose and measurements of the environment. In general, the motion model is derived from the kinematics of the robotic system, without taking the system dynamics into account. In this article, the authors investigate the performance and efficacy of standard SLAM algorithms when the dynamics of the robotic system is taken into account in the motion model and provide experimental results to complement the simulation findings.

Paper Details

Date Published: 3 April 2012
PDF: 8 pages
Proc. SPIE 8345, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2012, 83451F (3 April 2012); doi: 10.1117/12.914982
Show Author Affiliations
R. Jaai, Univ. of Maryland, College Park (United States)
N. Chopra, Univ. of Maryland, College Park (United States)
B. Balachandran, Univ. of Maryland, College Park (United States)
H. Karki, The Petroleum Institute (United Arab Emirates)


Published in SPIE Proceedings Vol. 8345:
Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2012
Masayoshi Tomizuka; Chung-Bang Yun; Jerome P. Lynch, Editor(s)

© SPIE. Terms of Use
Back to Top