issues
search
Roboy
/
gym-roboy
Pip package gym environment for deep roboy control
2
stars
1
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
_did_complete_successfully() method called with different inputs
#47
tomasruizt
closed
5 years ago
0
Update docs, restructure repo
#46
tomasruizt
closed
5 years ago
0
Remove constraint on tendon_vels
#45
tomasruizt
closed
5 years ago
0
Rename ROSProxy to SimulationClient
#44
tomasruizt
closed
5 years ago
5
Rename vel to vels and angle to angles
#43
tomasruizt
closed
5 years ago
0
Remove unused logging
#42
tomasruizt
closed
5 years ago
0
set new goal on env reset()
#41
tomasruizt
closed
5 years ago
3
Train script takes output dir
#40
tomasruizt
closed
5 years ago
0
Introduce maximum episode length, increase bonus for reaching goal
#39
tomasruizt
closed
5 years ago
0
split functionality of training and evaluation scripts
#38
tomasruizt
closed
5 years ago
0
Remove duplicate code
#37
ALPA-Industry-and-Technology
closed
5 years ago
2
Observation service
#36
BarisYazici
closed
5 years ago
0
Create new observation service
#35
BarisYazici
closed
5 years ago
1
Extra reward
#34
ALPA-Industry-and-Technology
closed
5 years ago
0
Extra reward
#33
ALPA-Industry-and-Technology
closed
5 years ago
0
Factor robot specific constants into MsjRobotState
#32
tomasruizt
closed
5 years ago
2
Feature parallel
#31
BarisYazici
closed
5 years ago
0
Remove ros_proxy delay & add load test for it
#30
tomasruizt
closed
5 years ago
0
Parallelization of gym_roboy
#29
BarisYazici
opened
5 years ago
1
Delete visualization related stuff
#28
BarisYazici
closed
5 years ago
1
Remove target publish
#27
BarisYazici
closed
5 years ago
0
Switch PPO1 with PPO2
#26
tomasruizt
closed
5 years ago
4
Exp reward
#25
tomasruizt
closed
5 years ago
0
Penalty for reaching unfeasible state
#24
tomasruizt
closed
5 years ago
1
New approach
#23
ALPA-Industry-and-Technology
closed
5 years ago
1
Add joint_angle_vel boundary and normalization
#22
tomasruizt
closed
5 years ago
0
BUGFIX: reward is lower with joint_vel_penalty
#21
tomasruizt
closed
5 years ago
0
Delete the rviz publish function
#20
BarisYazici
closed
5 years ago
0
Not calling reset() on unfeasible areas
#19
tomasruizt
closed
5 years ago
1
Reward based on joint_vel
#18
tomasruizt
closed
5 years ago
5
Script to run all tests
#17
tomasruizt
closed
5 years ago
1
Runtime constraints to reward range
#16
tomasruizt
closed
5 years ago
1
Enable Tensorboard and save to 'model' file
#15
tomasruizt
closed
5 years ago
2
Alex dev baris
#14
BarisYazici
closed
5 years ago
1
goal state is displayed in the origin in RVİZ
#13
BarisYazici
closed
5 years ago
4
Check the tensorboard logging/ Loss graph
#12
BarisYazici
closed
5 years ago
0
Small joint velocities (actually 0)
#11
BarisYazici
closed
5 years ago
2
The l2 distance for success is currently based on actions
#10
tomasruizt
closed
5 years ago
2
Action from agent is only -2.0 or 2.0
#9
BarisYazici
closed
5 years ago
1
New goal service, feasibility feature and dimensions bug solved
#8
BarisYazici
closed
5 years ago
6
Put a dependency of stable baselines. Clean the repo and clean the readme
#7
BarisYazici
closed
5 years ago
0
Folder rename fix to adapt OpenAI gym custom environment style
#6
BarisYazici
closed
5 years ago
0
Add an example script running PPO on our MsjEnv
#5
tomasruizt
closed
5 years ago
0
Printing robot_state parameters in a cleaner way
#4
BarisYazici
closed
5 years ago
0
Add tests for our entire MsjEnv class
#3
tomasruizt
closed
5 years ago
0
Add scripts to launch cardsflow & ROS2 bridge
#2
tomasruizt
closed
5 years ago
2
Add test for step function and refactor till it passes
#1
tomasruizt
closed
5 years ago
1