Courses:

Cognitive Robotics >> Content Detail



Projects



Projects

This section contains documents that could not be made accessible to screen reader software. A "#" symbol is use to denote such documents.

Students in 16.412 complete two projects. For the 'Advanced Lecture', students research a topic in the field of cognitive robotics, and deliver a lecture in-class on their selection. For the 'Final Project', teams expand the field through innovative methods or applications. Students report on the project through individual papers and a brief team presentation. All work is courtesy of the students named, and used with permission.



Advanced Lecture Information


The following advanced lecture summary is excerpted, in part, from the Advanced Lecture Proposal Guidelines (PDF)

Objectives

The purpose of the advanced lecture is to identify one to two advanced research methods, and to present these clearly and pedagogically, through a 45 minute oral presentation, through a written tutorial, and in the case of a three person team, with a demonstration. An additional objective is to learn to develop an understanding of the literature in a collaborative context, through one to two partners.

Advanced lectures improve the class' awareness of current research in the field of cognitive robotics. The instructors often re-use them in future semesters to keep the course up to date. A list of suggested topics is linked at the bottom of this section.

Grading

Advanced lectures are prepared by students and presented during scheduled lecture sessions. Students perform a 'dry-run' in advance of their in-class presentation. The dry-run is an opportunity for students to exchange feedback on their presentations. The revised lecture, delivered in-class, is subject to the following scoring scheme:

Scoring Guidelines for Technical Presentations (PDF)

Guidelines

In previous semesters, this project was introduced through a sequence of assignments: a warm-up literature review, and an advanced lecture proposal. Although the warm-up exercise was not assigned in Spring 2004, it introduces the advanced lecture project itself, which was undertaken by all students this semester.

Advanced Lecture Warm-Up Exercise (PDF)

Advanced Lecture Proposal Guidelines (PDF)

Advanced Lecture Submission Guidelines (PDF)

Suggested Topics and Readings for Advanced Lecture and Project (PDF)



2005 Advance Lectures



STUDENTSLECTURES
Lawrence Bush, Antonio Jimenez, and Brian BairstowSolving POMDPs Through Macro Decomposition (PDF - 3.7 MB)
Jennifer Novosad, Justin Fox, and Jeremie PoulyCognitive Game Theory (PDF - 2.8)
James Lenfestey, Thomas Temple, and Ethan HoweIntroduction to Probablisitic Relational Models (PDF)#
Kaijen Hsiao, Henry Lefebvre de Plinval-Salgues, and Jason MillerParticle Filters and their Applications (PDF - 1.3 MB)
Brian Mihok and Michael TerryStatistical Learning and Inference Methods for Reasoning in Games (PDF)
Thomas Coffee, Shannon Dong, and Shen QuHuman-Computer Interaction (PDF - 1.1 MB)



Final Project Information


The following final project summary is excerpted, in part, from the Final Project Guidelines (PDF)

Objectives

The purpose of the project is to develop a deep understanding of one or two methods for creating cognitive robots and intelligent embedded systems, and to innovate upon these methods, to lend novel insight into their behavior through analysis or to apply the method in an innovative manner.

More specifically, you should demonstrate the ability to:

  • Clearly state and motivate an interesting, focused innovation to intelligent embedded systems. An innovation may be an important analytical question, a novel algorithmic extension or an innovative application.
  • Extract and evaluate the relevant literature using the Web and library resources.
  • Provide a simple explanation for the algorithms used in your project, using pedagogical examples to highlight key features of the algorithm.
  • If a design project, describe the design of the intelligent embedded systems you are creating and the rationale for the method applied in the context of the project. If this is an analysis project, then described the experimental method that you are pursuing.
  • Implement and demonstrate an algorithm or application in support of your project goals.
  • Evaluate the approach analytically and/or empirically.

Grading

  • A - represents mastery: the ability to analyze and extend existing methods in a way that is novel and insightful; the ability to explain and motivate in a manner that is particularly intuitive.
  • B - represents solid competence: the ability to articulately motivate, explain, implement and evaluate a focused set (i.e., 1 or 2) of intelligent embedded systems methods.
  • C - represents partial competence of the above.

Report Guidelines

The report should reveal a depth of understanding. It should communicate the objectives, core description, developments and results of your project. Three main elements to present in your report:

  1. Articulation of the set of methods upon which your project is built, in a pedagogical (tutorial-like) manner.
  2. Empirical and/or analytical evaluation and insights.
  3. Innovation that involves applying and/or extending methods in a novel way.

Presentation Guidelines

Each team gives a presentation (approximately 5 minutes/group member). Selected student work from Spring 2005 and Spring 2004 are linked in the tables below. A list of suggested topics is linked at the bottom of this section.



2005 Final Projects



STUDENTSTITLESPROPOSALSFINAL PROJECTS
Lawrence Bush, Antonio Jimenez, and Brian BairstowAdaptable Mission Planning for Kino-Dynamic Systems(PDF)(PDF)
Thomas Coffee, Shuonan Dong, and Shen QuSpatial Intent Recognition Using Optimal Margin Classifiers(PDF)(PDF)
Ethan Howe and Jennifer NovosadExtending SLAM to Multiple Robots
Kaijen Hsiao, Henry Lefebvre de Plinval-Salgues, and Jason MillerMapping Contoured Terrain: A Comparison of SLAM Algorithms for Radio-Controlled Helicopters(PDF)(PDF - 1.1 MB)
James LenfesteyA SIFT-based Pictorial Image Model
Brian Mihok and Michael TerryA Bayesian Net Inference Tool for Hidden State in Texas Hold'em Poker(PDF)(PDF)
Jeremie Pouly and Justin FoxAn Empirical Investigation of Mutation Parameters and Their Effects on Evolutionary Convergence of a Chess Evaluation Function(PDF)(PDF)
Thomas TempleGPS Integrity Monitoring(PDF)



2004 Final Projects



STUDENTSPROJECTS
Alexander OmelchenkoAutonomous Visual Tracking Algorithms

Presentation
Report
Lars Blackmore and Steve BlockCooperative Q Learning

Presentation
Report by Block
Report by Blackmore
Seung H. Chung, Robert T. Effinger, Thomas Léauté, and Steven D. LovellModel-based Programming for Cooperating Vehicles

Presentation
Report by Léauté (PDF)
Report by Chung (PDF)
Report by Effinger (PDF)
Report by Lovell
Morten Rufus Blas and Søren RiisgaardSLAM for Dummies

Presentation (PDF - 1.5 MB)
Report (PDF)#
Personal comments by Blas (PDF)
Personal comments by Riisgaard (PDF)
Vikash K. MansinghkaTowards Visual SLAM in Dynamic Environments

Presentation (PDF)
Report (PDF)

 








© 2017 CourseTube.com, by Higher Ed Media LLC. All Rights Reserved.