Progress for week 13 (2015)
From Robin
(Difference between revisions)
(→Else-Line) |
(→Stian) |
||
Line 14: | Line 14: | ||
=== Accounting === | === Accounting === | ||
+ | * Wrote. | ||
+ | * Problem with home disk dropping files and preventing writing of results (!). | ||
+ | * Lots of improved Matlab scripts. | ||
+ | * Nearly finished training set size testing. | ||
+ | * Nearly finished objective testing. Could need some more runs on particular sets. | ||
+ | * Started new prefilter testing. | ||
+ | ** Implemented some new filters. | ||
+ | * Tested different runtime measurements and their stability. | ||
+ | * Created and assembled stereo camera mount | ||
+ | ** Experimented with camera parameters suitable for stereo | ||
== Snorre == | == Snorre == |
Revision as of 08:35, 13 April 2015
Contents |
Student template (copy this for your entry)
Budget
- Todo 1
- Todo 2
Accounting
- Done 1
- Done 2
Stian
Budget
- Write
- Plan further experiments
Accounting
- Wrote.
- Problem with home disk dropping files and preventing writing of results (!).
- Lots of improved Matlab scripts.
- Nearly finished training set size testing.
- Nearly finished objective testing. Could need some more runs on particular sets.
- Started new prefilter testing.
- Implemented some new filters.
- Tested different runtime measurements and their stability.
- Created and assembled stereo camera mount
- Experimented with camera parameters suitable for stereo
Snorre
Budget
- Finish writing about walking methods
- Do simulations on uneven terrain
Accounting
Rune
Budget
- Perform new simulations with updated pathfinding
- Write
Accounting
Bendik
Budget
- test enemy avoidance strategies
- Improve simulation tool
- Write about simulation tool
Accounting
Eivind
Budget
- Keep modelling and printing parts
- AI testing
- Update control-program for the new AI
- Write at least 6 pages
Accounting
Else-Line
Budget
- 45 pages total
- Continue testing on real robot
- Plot results
Accounting
- 60ish pages total
- Started evaluating gaits on physical robot, but then a core part broke. New one printed and attached
- Plotted and replotted results
- Ran duplicated tests in simulation to achive statistical significance
- Also ran test with local search only, performs worse than both the MA and the plain GA