-

hey viewer, we're moving!

We are currently transitioning to a new web system, so we are not updating this wikisite anymore.

The public part of the new web system is available at http://www.ira.disco.unimib.it


2012 Bayes Filtering, with application to Robotic Perception

From Irawiki

Revision as of 13:47, 24 June 2013 by IraWikiSysop (Talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Contents

Short description of the course

The course gives an introduction to the techniques known implementing Bayes Filtering, and presents an application of such techniques to robotic perception. The course has an hands-on approach, which requires some effort during the course, though not after.

Lecturer(s)

Schedule

  • Wednesday May 2nd, 2012; 08:30 - 10:30; U14-T014, lecture 1
Basics of probabilities
Basics of Bayes Filtering
  • Monday May 7th, 2012; 08:30 - 10:30; U14-T023, lecture 2
Gaussian Filtering (Kalman)
Extended Kalman Filter
  • Thursday May 10th, 2012; 08:30 - 11:30; Sala Lettura piano 1, lab1
laboratory: complexity analysis of Kalman Filtering
  • Monday May 14th, 2012; 08:30 - 10:30; U14-T023, lecture 3
Jacobian-based propagation of uncertainty
EKF for parameter estimation with an implicit output transform
Unscended Kalman Filter
  • Wednesday May 16th, 2012; 08:30 - 10:30; U14-T014, lecture 4
Information Form Gaussian Filtering
Non parametric filtering
  • Thursday May 17th, 2012; 09:30 - 12:30; Sala Lettura piano 1, lab2
laboratory: development of the system simulator for a simple target tracking example
  • Monday May 21st, 2012; 08:30 - 10:30; U14-T023, lecture 5
Non parametric filtering
  • Tuesday May 22th, 2012; 09:30 - 12:30; Sala Lettura piano 1, lab3
laboratory: development of EKF for a simple target tracking example
  • Wednesday May 23th, 2012; 08:30 - 10:30; U14-T014, lecture 6
Motion and sensor modeling
  • Monday May 28th, 2012; 08:30 - 10:30; U14-T014, lecture 7
Introduction to the SLAM problem
  • Thursday May 31th, 2012; 09:30 - 12:30; Sala Lettura piano 1, lab4
laboratory: development of EKF for a simple target tracking example, conclusion
  • Tuesday june 5th, 2012; 09:30 - 12:30; Sala Riunioni piano 1, lab5
laboratory: development of PF for a simple target tracking example
  • Thursday June 7th, 2012; 09:30 - 12:30; U14-T024, lab6
laboratory: development of PF for a simple target tracking example, conclusion

Grading

Grading will be based on the evaluation given by the lecturers on the quality of:

  • the documentation prepared by students on specific course topics;
  • the programs (typ. mlab), and the comments therein, the students will develop during the labs.

The grading scale adopted is:

A    90% and above (Excellence)     BAND 6
B    80-89%        (Very Good)      BAND 5
C    70-79%        (Good)           BAND 4
D    60-69%        (Average)        BAND 3
E    50-59%        (Unsatisfactory) BAND 2
F    49% and under (Failure)        BAND 1

Course Material

The course material, beside the reference below, is available here; login and password have been communicated during classes. Notice though that most of the "Introduction to Bayes Filtering" part is perfectly covered by the first chapters of the book "Probabilistic Robotics".

Registration

Please send an email to sorrenti at disco dot unimib dot it stating that you would like to participate to the course; please specify which part(s) of the course you are going to attend.

ECTS notes

Here is our best estimate of the effort required to pass the course by the "average student". This estimate is provided for the students that need an estimate of the effort to get the credits. This section needs a revision!

  • Introduction to Bayes filtering
  • actual class hours: 10
  • individual study for the class topics: 20
  • actual lab. hours: 15
  • individual working on the lab. assignments: 15
  • total: 10 + 20 + 15 + 15 = 60
  • Robotic application
  • actual class hours: 4
  • individual study for the class topics: 8
  • actual lab. hours: 3
  • individual working on the lab. assignments: 3
  • total: 4 + 8 + 3 + 3 = 18

Attendance

to be verified, as for 23.05.12
                       02.05  07.05  10.05  14.05  16.05  17.05  21.05  22.05  23.05  28.05  31.05  05.06  07.06
                       lect1  lect2  lab1   lect3  lect4  lab2   lect5  lab3   lect6  lect7  lab4   lab5   lab6 
Ali Hashim             -      -      -      -      -      ok     ok     ok     ok     ok     ok     ok          
Citrolo Andrea         -      -      ok     ok     ok     ok     ok     ok     -      -      ok     ok          
Cucci Davide           -      ok     ok     -      -      -      -      -      -      -      -      ok          
Luciani Davide         -      ok     ok     ok     ok     -      ok     ok     ok     ok      -     ok          
Morselli Alessandro    ok     ok     ok     ok     ok     ok     -      ok     -             ok     ok          
Fontana Simone         ok     ok     ok     -      ok     ok     -      ok     ok     -      ok     ok          
Mazzini Andrea                                                                                      ok          
Prisciantelli Lucia                                                                                 ok          
Sacchi Francesco                                                                                                
Ventimiglia Giuseppe                                                                                ok          

Grades

dns = did not submit, na = not applicable;


student final result lab1 complexity analysis of KF lab2, simulator lab3, EKF lab4, PF
Ali Hashim final (whole course) not submitted (will submit by 24th night) submitted part 1 not submitted not submitted
Citrolo Andrea final (whole course) submitted submitted part 1 not submitted not submitted
Cucci Davide final (whole course): A- submitted, B submitted (given Davide's background the task has been changed in "analysis of capabilities of UKF vs. EKF in handling linearizations, from Julier's paper"), A submitted, code B+, comments B, overall B+
Luciani Davide final (whole course) submitted submitted part 1 not submitted not submitted
Morselli Alessandro final (whole course): A+ submitted, A+ submitted, A+, given to other students as reference submitted, code A+, comments A, overall A+ submitted, code A+ (minor: why the length of the life of the mobile feature has been randomized?), comments A, overall A+
Fontana Simone final (whole course) submitted submitted part 1 not submitted not submitted

Commentary

Please see the same section concerning the previous instance of the course.

Personal tools