Progress for week 40 (2014)

Fra Robin

(Forskjeller mellom versjoner)
Gå til: navigasjon, søk
(Eivind)
 
(18 mellomrevisjoner ikke vist.)
Linje 10: Linje 10:
== Else-Line ==
== Else-Line ==
=== Budget ===
=== Budget ===
 +
* Attend second part of simulation system introduction
* Experiment with simulations
* Experiment with simulations
-
* Implement a simple local search (simulated annealing?)
+
* Implement a simple local search (random walk, simulated annealing?)
=== Accounting ===
=== Accounting ===
 +
* Attended second part of workshop
 +
* Experimented a little with simulations, starting to understand how things work
== Stian ==
== Stian ==
=== Budget ===
=== Budget ===
* Further focus on software algorithm comparisons.
* Further focus on software algorithm comparisons.
 +
** Try a GA/SA approach to algorithm parameter optimization using one of the current stereo frameworks to evaluate the results.
=== Accounting ===
=== Accounting ===
 +
* Started building a framework around the new middlebury SDK.
 +
** Delayed somewhat as I was unable to replicate the official results on a given testset.
 +
 +
== Bendik ==
 +
=== Budget ===
 +
* Continue working on the simulation tool in Python
 +
* Plan communication between AI and Control system
 +
 +
=== Accounting ===
 +
 +
== Eivind ==
 +
=== Budget ===
 +
* Finish implementing UI
 +
* Decide if new motors must be acquired
 +
 +
=== Accounting ===
 +
* Scrapped UI
 +
* Researched new motors
 +
 +
== Rune ==
 +
=== Budget ===
 +
* Implement a basic path-planning algorithm for testing in roborobo.
 +
* Read more papers on behavior prediction.
 +
 +
=== Accounting ===
 +
*Studied c++
 +
*Created roborobo shell
 +
 +
== Snorre ==
 +
=== Budget ===
 +
* Write and experiment with local modules on the robot (C++)
 +
* Participate the second simulator workshop
 +
* Look at some Evolutionary algorithms for testing
 +
 +
=== Accounting ===
 +
* Participated the second workshop
 +
* Experimented with different walking gaits, in addition to randomising some parameters
 +
* Started using the old version of the SDK, and it fixed a lot of problems
 +
* Looked at some Webots tutorials
 +
* Started reading the paper about NSGA-II
 +
 +
== André ==
 +
=== Budget ===
 +
* Implement the odometry calculation in python
 +
* Look at the communication protocol to be used
 +
* Try to find some off the shelf laser and bacons

Nåværende revisjon fra 14. okt 2014 kl. 09:22

Innhold

Student template (copy this for your entry)

Budget

  • Todo 1
  • Todo 2

Accounting

  • Done 1
  • Done 2

Else-Line

Budget

  • Attend second part of simulation system introduction
  • Experiment with simulations
  • Implement a simple local search (random walk, simulated annealing?)

Accounting

  • Attended second part of workshop
  • Experimented a little with simulations, starting to understand how things work

Stian

Budget

  • Further focus on software algorithm comparisons.
    • Try a GA/SA approach to algorithm parameter optimization using one of the current stereo frameworks to evaluate the results.

Accounting

  • Started building a framework around the new middlebury SDK.
    • Delayed somewhat as I was unable to replicate the official results on a given testset.

Bendik

Budget

  • Continue working on the simulation tool in Python
  • Plan communication between AI and Control system

Accounting

Eivind

Budget

  • Finish implementing UI
  • Decide if new motors must be acquired

Accounting

  • Scrapped UI
  • Researched new motors

Rune

Budget

  • Implement a basic path-planning algorithm for testing in roborobo.
  • Read more papers on behavior prediction.

Accounting

  • Studied c++
  • Created roborobo shell

Snorre

Budget

  • Write and experiment with local modules on the robot (C++)
  • Participate the second simulator workshop
  • Look at some Evolutionary algorithms for testing

Accounting

  • Participated the second workshop
  • Experimented with different walking gaits, in addition to randomising some parameters
  • Started using the old version of the SDK, and it fixed a lot of problems
  • Looked at some Webots tutorials
  • Started reading the paper about NSGA-II

André

Budget

  • Implement the odometry calculation in python
  • Look at the communication protocol to be used
  • Try to find some off the shelf laser and bacons
Personlige verktøy