Progress for week 4 (2015)
From Robin
(Difference between revisions)
(→André) |
|||
(6 intermediate revisions not shown) | |||
Line 25: | Line 25: | ||
=== Accounting === | === Accounting === | ||
+ | * Attempted to compile OpenCV with the ximgproc module enabled. | ||
+ | ** ximgproc (extended image processing) contains several new and relevant filtering techniques. | ||
+ | * Started experiments with Gaussian and Bilateral filters. | ||
== Else-Line == | == Else-Line == | ||
Line 32: | Line 35: | ||
=== Accounting === | === Accounting === | ||
+ | * Added more constraints on genes, some improvement in walking patterns | ||
+ | ** Limits on amp-off-phase, special case for "knee" joints, can not have negative offset | ||
+ | ** Offset body always set to 0 | ||
+ | ** Mirroring around body | ||
+ | * Wrote down some more notes | ||
== Snorre == | == Snorre == | ||
Line 38: | Line 46: | ||
* Make transferability function | * Make transferability function | ||
* Write on thesis | * Write on thesis | ||
+ | === Accounting === | ||
+ | * Upgraded the robot to the newest version | ||
+ | * Did simulations to use for transferring | ||
+ | * Modified a controller program to control the physical robot | ||
+ | * Wrote some notes | ||
== Rune == | == Rune == | ||
Line 46: | Line 59: | ||
=== Accounting === | === Accounting === | ||
- | + | *Learning component implemented!27. jan 2015 kl. 09:44 (UTC)~ | |
+ | *Investigated options for learning by observing. | ||
== Bendik == | == Bendik == | ||
Line 56: | Line 70: | ||
=== Accounting === | === Accounting === | ||
+ | * Research optimizations of theta* | ||
+ | * Build gripper | ||
== André == | == André == |
Current revision as of 10:01, 27 January 2015
Contents |
Student template (copy this for your entry)
Budget
- Todo 1
- Todo 2
Accounting
- Done 1
- Done 2
Eivind
Budget
- Decide on a solution for the front grippers, 3d-model and print parts.
- Do realistic testing on the game-board with AI
- Research solutions for reliably run the robot on battery power.
- Help test beacon positioning system
Accounting
- Front grippers modelled and ready for printing
- Researched robot wheels
Stian
Budget
- Implement some prefilters.
- Start redoing some of the experiments with various prefiltering enabled.
Accounting
- Attempted to compile OpenCV with the ximgproc module enabled.
- ximgproc (extended image processing) contains several new and relevant filtering techniques.
- Started experiments with Gaussian and Bilateral filters.
Else-Line
Budget
- Continue improving genes
- Write some more about background, plus maybe some about results so far
Accounting
- Added more constraints on genes, some improvement in walking patterns
- Limits on amp-off-phase, special case for "knee" joints, can not have negative offset
- Offset body always set to 0
- Mirroring around body
- Wrote down some more notes
Snorre
Budget
- Do experiments
- Make transferability function
- Write on thesis
Accounting
- Upgraded the robot to the newest version
- Did simulations to use for transferring
- Modified a controller program to control the physical robot
- Wrote some notes
Rune
Budget
- Finish implementation of learning component.
- Scenario (behavior/paths).
- Look at options for learning by observations.
Accounting
- Learning component implemented!27. jan 2015 kl. 09:44 (UTC)~
- Investigated options for learning by observing.
Bendik
Budget
- Write 10 pages
- Research optimizations of theta*
- Build gripper
- Score 1 point!
Accounting
- Research optimizations of theta*
- Build gripper
André
Budget
- Write 10 pages of notes
- finish the gripper
- Start looking at combining the position from encoders with the beacons.
Accounting
- Written a couple of pages
- 3D designet the tower for the beacon
- Looked at the gripper, and we think we have come up with a good solutions (inspired by youtube)