Abstract
Planning can be used in a variety of applications. In this paper we will discuss those planning techniques that Apply to the task of robotic path planning - Here a planner is used to generate "paths" which a robot can follow to maneuver from some point A to another point B, while at the same time avoiding all obstacles. All approaches discussed in this paper are based on viewing the robot as a sphere . By assuming this , the need to consider the robot's orientation as it moves along a proposed path is eliminated . Another requirement is that not only must a successful path be found, but this path should also be the shortest path through the space . Since "finding the shortest path between two points that avoids a collection of poly-hedral obstacles in three dimensions is already computationally intractable" and 3-D robotic vision may not be available, the discussion in this paper will be restricted to a 2D plane (this infers that the robot's terrain is a flat hard surface). Object recognition will also not be considered, only the ability to determine that there is some object present (whether it 's a table, chair or T. V. doesn't matter ) . Its length and width must be known or determined. The height of the object is not important as the robot will go around the object and not under or over it (can only obtain height information from a 3D plane) . To simplify the overall problem domain we assume that obstacles are not in motion (IE, the objects are not in constant motion; objects can be moved to new stationary locations and new paths around them searched for). The discussion will also restrict the degrees of freedom of the robot to 2. This is again done to reduce the complexity of the domain. As more degrees of freedom are considered, the path planning problem becomes increasingly complex. Finally, we will assume the robot's velocity remains constant (again to reduce the complexity of the domain).
Library of Congress Subject Headings
Mobile robots; Robots--Motion; Robots--Programming; Robot vision
Publication Date
1993
Document Type
Thesis
Department, Program, or Center
Computer Science (GCCIS)
Advisor
Wolf, Walter
Advisor/Committee Member
Kazemian, Feredoun
Advisor/Committee Member
Biles, Al
Recommended Citation
Switzer, Barbara T., "Robotic path planning with obstacle avoidance" (1993). Thesis. Rochester Institute of Technology. Accessed from
https://repository.rit.edu/theses/789
Campus
RIT – Main Campus
Comments
Note: imported from RIT’s Digital Media Library running on DSpace to RIT Scholar Works. Physical copy available through RIT's The Wallace Library at: TJ211.415 .S94 1993