Proposal Number: 16
Should be considered for best Project Proposal: No
Organization/Clarity of the proposal: 9
Overall Project Idea: 6
        I think that this proposal has a number of vague points that are likely just improperly transferred from the mind of the author to word.  I guess I will first try and put
to word what I understand the project to be and what principles are to be explored throughout the project.  This project is supposed to be an exploration of an artificial being’s
ability to create a proper representation of a working environment.  This is clearly a necessity for robots to be able to work in a household setting.  The proposed approach for
this is the embedding of RFID tags that are associated with proper affordances.  I find this to be already a bit of a stretch on reality as the field has some inherent need for
robots that can perform tasks in a natural environment.  This concept of nature is betrayed by the placement of RFID tags.  As stated in the proposal, however, such modified
environments have been created many times before for similar purposes.
        The modified approach seems like a clever idea to implement a potentially complex system.  Setting up a system of RFID tags, readers, and some sort of association is
likely too complex a task to implement in the rather small time frame supplied for this project.  The two simplifications of adding QR codes to replace RFID tags and of using a
wizard of oz system in stead of an actual robotic platform are both no doubt appropriate and effective methods of simplifying the system without losing the real meaning of the
project.
        As I understand the project, a human being will move around a room that is implanted with the QR codes, taking inventory of the locations and values of the codes.  This
inventory is then encoded and a model is built, and some recall metric is used to test the model for accuracy.  As I understand, these are the three modes discussed in the
proposal.
        I find the proposal itself very well organized and flows properly from motivation to implementation, and finally to evaluation and conclusion.  The individual steps or the
projects are well stated and the project plan is quite easy to understand (at least in terms of its chonology).
        There are a few points that I’m not quite sure about after reading the proposal.  The evaluation section does state the given goals of each mode of the project.  Those
goals do, however, lack a certain precision.  The stated goals have some implication of a perfection requirement with statements like “the robot is expected to respond to a seroes
of commands....“  There should be some expectation of imperfection as such is present in any system.  Some leeway should be considered for the final evaluation of the project.
        Some of the details of some of the modes also seem to be missing.  I’m not certain what sort of commands the system is expected to process.  This is further confusing
considering the use of a “Wizard of Oz” system.  Simply put, how is a human supposed to properly represent the limited representation of the proposed robot.
        I’m also curious as to the scope of the system and the size of the room or functional area that this proposal talks about.  There is mention of doors so its likely a
system that will be implemented in an apartment or at least a number of rooms.  Furthermore, I’d like to know how many such tags are to be placed and on how many items these tags
will be placed.  I’d also like to know what kind of test information will be embedded into the tags themselves.
        The change from RFID tags to those requiring direct line of sight imposes a notable change on the system itself.  In the ideal proposed system, the robot need only be
somewhere near a tag to extract pertinent information.  The modified system will, however, require direct line of sight to the individual tags making it somewhat more difficult to
properly locate all the tags.  I tend to think that this difference is quite difficult to properly account for.
        I rather like the irony of using an Android phone to simulate a robotics platform.  The idea of using a commercially avaialable mobile device is, itself, quite appealing. 
A large number of projects for this class will no doubt use robotic platforms that are largely unfeasible to replicate in an environment outside of the academic world.
        I also find that this proposal is missing an aspect of a source of personal interest.  The interest and potential gain to society in general is quite obvious, but the
assignment itself was meant as an exploration of a personal interest in developmental robotics and the source of personal interest in this project was not stated.
        As I understand the concept, the android system is not sufficient to emulate a robotic platform.  Simply put, the phone lacks actuators.  The need for embodiment is very
necessary for proper development and the android system breaks does not meet this requirement.  Although this does not really ruin the purpose of the project itself, it makes the
title somewhat misleading.  The phone is not being used as a robotics platform but, perhaps, is a sensor of a robotic platform that, at this stage, lacks actuators.
        The difficulty of this project seems appropriate and likely well within the bounds of the author.  All of the seemingly appropriate skills were mentioned as a strength of
the author.  At the same time, this project seems sufficiently difficult for the project assignment and the time period assigned to complete the project.