Progress for week 3 (2015)
From Robin
(Difference between revisions)
(→Budget) |
(→André) |
||
(13 intermediate revisions not shown) | |||
Line 7: | Line 7: | ||
* Done 1 | * Done 1 | ||
* Done 2 | * Done 2 | ||
+ | |||
+ | == All students == | ||
+ | === Accounting === | ||
+ | * Gingerbread house vs X2 robot | ||
+ | * http://youtu.be/ISnUJ6OQoLw | ||
== Eivind == | == Eivind == | ||
Line 16: | Line 21: | ||
=== Accounting === | === Accounting === | ||
+ | * Modified lift to work with old gears | ||
+ | * AI can now control the lift | ||
+ | * Implemented multilevel logging functionality | ||
+ | * Program stress-testing and improvements | ||
+ | * Researched proximity sensors | ||
== Else-Line == | == Else-Line == | ||
Line 24: | Line 34: | ||
=== Accounting === | === Accounting === | ||
+ | * Tried adding different constraints to the genes, without making much improvement | ||
+ | * Read article on learning and evolution on neural networks | ||
== Stian == | == Stian == | ||
Line 32: | Line 44: | ||
=== Accounting === | === Accounting === | ||
+ | * Read several filtering papers. | ||
+ | * Made som thesis notes. | ||
+ | * Some code restructuring for easier filter experiments. | ||
+ | * Rerunning simulations. | ||
+ | |||
+ | == André == | ||
+ | === Budget === | ||
+ | * Finish the beacon system. borrowed aluinium tape from Stian that may work | ||
+ | * Test the beacon system, and calculate the errors so that the angles will be as correct as possible | ||
+ | |||
+ | === Accounting === | ||
+ | * Almost finished the beacon systems. Some more testing and callibrating with 3 beacons | ||
+ | |||
+ | == Rune == | ||
+ | === Budget === | ||
+ | * Decide on method/representation and start implementing ML. | ||
+ | * Scenario (behavior/paths) | ||
+ | * Write | ||
+ | |||
+ | === Accounting === | ||
+ | *Decided on and started implementation of ANN with Q-learning. | ||
+ | *Began writing (headlines/filling in). | ||
+ | |||
+ | == Snorre == | ||
+ | === Budget === | ||
+ | * Do experiments | ||
+ | * Make transferability function, motion capture? | ||
+ | * Write on thesis | ||
+ | === Accounting === | ||
+ | * Done some experiments, tweaked the controller system | ||
+ | |||
+ | == Bendik == | ||
+ | === Budget === | ||
+ | * Finish AI simulator | ||
+ | * Research and implement theta* | ||
+ | * Find thesis problem | ||
+ | * Write on thesis | ||
+ | * Plan gripper | ||
+ | |||
+ | === Accounting === | ||
+ | * Finished AI simulator | ||
+ | * Implemented theta* found and read 3 relevant articles | ||
+ | * No problem! | ||
+ | * found solution for gripper |
Current revision as of 09:08, 22 January 2015
Contents |
Student template (copy this for your entry)
Budget
- Todo 1
- Todo 2
Accounting
- Done 1
- Done 2
All students
Accounting
- Gingerbread house vs X2 robot
- http://youtu.be/ISnUJ6OQoLw
Eivind
Budget
- 3d-print new gears for the lift stepper-motor.
- Test the new position-control system together with AI
- Look into solutions for logging program output
- Purchase short-range proximity sensors (ultrasonic/infrared)
Accounting
- Modified lift to work with old gears
- AI can now control the lift
- Implemented multilevel logging functionality
- Program stress-testing and improvements
- Researched proximity sensors
Else-Line
Budget
- Check crossover method, as results were strange
- Improve ExampleGenes? For better walking patterns
- Read article on learning and evolution
Accounting
- Tried adding different constraints to the genes, without making much improvement
- Read article on learning and evolution on neural networks
Stian
Budget
- Arrange meeting with Omnivision
- Plan thesis
- Continue prefiltering experiments. Rerun simulations.
Accounting
- Read several filtering papers.
- Made som thesis notes.
- Some code restructuring for easier filter experiments.
- Rerunning simulations.
André
Budget
- Finish the beacon system. borrowed aluinium tape from Stian that may work
- Test the beacon system, and calculate the errors so that the angles will be as correct as possible
Accounting
- Almost finished the beacon systems. Some more testing and callibrating with 3 beacons
Rune
Budget
- Decide on method/representation and start implementing ML.
- Scenario (behavior/paths)
- Write
Accounting
- Decided on and started implementation of ANN with Q-learning.
- Began writing (headlines/filling in).
Snorre
Budget
- Do experiments
- Make transferability function, motion capture?
- Write on thesis
Accounting
- Done some experiments, tweaked the controller system
Bendik
Budget
- Finish AI simulator
- Research and implement theta*
- Find thesis problem
- Write on thesis
- Plan gripper
Accounting
- Finished AI simulator
- Implemented theta* found and read 3 relevant articles
- No problem!
- found solution for gripper