RAFS - Release 1
Release 1
In the first release, we are planning to get the robot to wander around the
room aimlessly (without running into anything), and beep when it recognizes
a chair. We are also going to implement some movement with the chair. This will
basically consist of the chair already being grasped (manually), with an object
positioned in front of the robot and chair. We are going to try to get the robot
to move around the chair. Peter will lead the work on the color recognition
using image processing, and J.D. will help him. Brad will head the research
in getting the robot to randomly move around the room, with Matt assisting him.
We will also need a contingency plan for robot complications. Although we will
start the research for localization in this release, it will not be implemented
until Release 2. A schedule of each person's task for this release is as follows (Brad - Green,
Matt - Red, Peter - Gray, J.D. - Blue, Entire Group - White):
Below is a description of each point in the chart. Some of the areas have a
lot of detail. To see this detail, simply click on the title.
Contingency Plan
Due to last semester's re-boot problem, it became evident that we needed to
set a back-up plan in case a similar situation should arise this semester.
Movement
This task deals with robot movement and wandering behavior within the Engineering
Building, room 2029. This module will enable the robot to randomly wander around
EB 2029 while the object recognition module executes in the background. When
an object (chair) is found, movement will halt and object recognition will take
control.
Object Recognition
As defined in our Project Design Document, we will be relying heavily on image
processing techniques to determine the placement of chairs in EB2029. Using
the ActivMedia Color Training Software (ACTS), we will locate chairs while roaming
about the room. This technology, combined with the sonar device of the robot,
will enable the robot to approach the chair and log its position.
Chair Movement
There are many problems that we see with the robot moving while a chair is
gripped. In this phase, we plan on getting familiar with these problems so in
subsequent phases we are not overwhelmed by these problems. For this phase,
we will assume the chair is gripped prior to movement. We will move around a
fixed object with the chair gripped. This is only to see how the robots movement
will be affected under these conditions. We would like to see how the sonar,
laser, camera, and movement will be affected.
Website
The website is a major part of our project. It will be informing everybody
about the progress of our project. It will continuously need to be updated throughout
the semester.
Integrate Modules (Movement & O.R.)
During the module integration phase of the project, we will be putting together
the two halves consisting of movement and object recognition. This will be the
most significant final step in building a complete software package. All modules
will be compiled and built together and then tested as a complete system.
Localization Research
Localization research involves two parts. Object recognition is the first step,
as the robot must keep record of where the found chairs are located. Second,
localization is important because the robots movement is skewed when it is traveling
with a large chair. The research will include these two sub-fields and also
all generic research such as reading manuals and also studying and building
code examples.
Release 2 Planning
The planning for Release 2 is already underway. We will be expanding the items
mentioned in the Release 2 Gantt chart into greater detail and also further
define the subtasks.