Information Technology Reference
In-Depth Information
Chapter 11
Path-Planning for Visual Servoing: A Review
and Issues
Moslem Kazemi, Kamal Gupta, and Mehran Mehrandezh
Abstract.
In this survey we provide a comprehensive technical review of existing
major approaches to path-planning for visual servoing. Visual servoing has been
introduced as a promising approach for sensor-based robotic tasks. The basic vi-
sual servoing task is to guide the motion of a robot with respect to a target ob-
ject based on the feedback obtained through a vision system. Amalgamation of
path-planning techniques with reactive visual servoing strategies can robustify ex-
isting image-based tracking systems in robotics applications where a high disparity
between the initial and desired views of a target is inevitable (
e.g.
, target intercep-
tion, space docking, reaching and grasping,
etc
). The planning stage does so by ac-
counting for critical constraints and uncertainties in the system resulting in a more
robust visual servoing process. We discuss different planning approaches, explain
the associated set of constraints and assumptions, and discuss the underlying path-
planning techniques along with the issues regarding their integration with reactive
visual servo controllers.
11.1
Introduction
The role of vision as a sensor for autonomous machines to interact with complex,
unknown, and dynamic environments is paramount. Visual servoing has been intro-
duced as a promising approach for sensor-based robotic tasks such as positioning
a robot with respect to a target and tracking a moving target via estimating its 3D
motion,
i.e.
egomotion analysis using vision. The basic visual servoing task is to
Search WWH ::
Custom Search