AndrejOrsula drl_grasping issues

AndrejOrsula / drl_grasping

Deep Reinforcement Learning for Robotic Grasping from Octrees

https://arxiv.org/pdf/2208.00818

BSD 3-Clause "New" or "Revised" License

378 stars 50 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Deploying on Franka Research 3 (real)

#111 lif314 closed 10 months ago
2
Training Error & Docker Tips

#108 lucarei closed 1 year ago
6
How much SDRAM is needed to reproduce the results in paper?

#105 YantianZha closed 1 year ago
0
Gazebo Source Installation

#104 lucarei closed 1 year ago
6
Pyoctree Attribute Error

#103 lucarei closed 1 year ago
2
Caffe Installation

#102 lucarei closed 1 year ago
1
Adding static Objects to the Reach Task

#100 Nils-ChristianIseke closed 2 years ago
2
Visualizing Reach Tasks

#99 Nils-ChristianIseke closed 2 years ago
2
Error while running training bash script

#98 Shanki5 closed 2 years ago
2
Kinova arm with SAC Algorithm not learning

#97 dsriaditya999 closed 2 years ago
4
Error running ex_train.bash in a headless way

#96 dsriaditya999 closed 2 years ago
3
Progress towards 1.1.0 release

#95 AndrejOrsula closed 2 years ago
0
Update Dockerfile

#94 AndrejOrsula closed 2 years ago
0
Add default SDF world for RL environments

#93 AndrejOrsula closed 2 years ago
0
Add support for Kinova Gen2 (j2s7s300) robot model

#92 AndrejOrsula closed 2 years ago
0
Replace fork of Gym-Ignition with upstream release (LocalCache)

#91 AndrejOrsula closed 6 months ago
1
`ex_enjoy_pretrained_agent.bash` hanging in fresh Docker image

#90 playertr closed 3 years ago
4
`ocnn` dependency cannot be built using `colcon`

#89 lisheng-han closed 3 years ago
2
O-CNN repository for local installation

#88 movefasta closed 3 years ago
11
FidgetSpin task

#87 AndrejOrsula opened 3 years ago
0
Update documentation with more comprehensive instructions

#86 AndrejOrsula closed 3 years ago
0
Refactor drl_grasping module to simplify usage and modifications

#85 AndrejOrsula closed 2 years ago
1
Add pre-trained agents that can be directly enjoyed without any training

#83 AndrejOrsula closed 3 years ago
0
Move all animations from README to another repository and purge them from this repo's history

#82 AndrejOrsula closed 3 years ago
1
Update docs

#81 AndrejOrsula closed 3 years ago
0
Update hyperparameters

#80 AndrejOrsula closed 3 years ago
0
Add UR5 with RG2 gripper to supported robots

#79 AndrejOrsula closed 3 years ago
0
Add instructions for examples in README

#78 lorepieri8 closed 3 years ago
2
Add option to use a separate network for stacked octrees

#77 AndrejOrsula closed 3 years ago
0
Add initial animations of grasp and domain randomization to README

#76 AndrejOrsula closed 3 years ago
0
Separate instructions for local installation and Docker

#75 AndrejOrsula closed 3 years ago
0
Document more details

#74 AndrejOrsula closed 3 years ago
0
Add scripts for interacting with model dataset from Fuel (training+testing)

#73 AndrejOrsula closed 3 years ago
0
Add option to preload replay buffer with demonstration

#72 AndrejOrsula closed 3 years ago
0
Octree with proprioceptive observations

#71 AndrejOrsula closed 3 years ago
0
Add few trained agents that can be enjoyed

#70 AndrejOrsula closed 3 years ago
1
Update documentation

#69 AndrejOrsula closed 3 years ago
1
Initial tuning of hyperparameters

#68 AndrejOrsula closed 3 years ago
0
Add basic runtime for real world evaluation on Franka Emika Panda

#67 AndrejOrsula closed 3 years ago
1
Add stacked octree observations

#66 AndrejOrsula closed 3 years ago
2
Add task- and workspace-based curriculum for Grasp tasks

#65 AndrejOrsula closed 3 years ago
0
Minor fixes/updates to Tasks(shaped reward), Randomizer and Docker

#64 AndrejOrsula closed 3 years ago
0
Add additional terminations that result in failure

#63 AndrejOrsula closed 3 years ago
2
Curriculum learning for grasping environment

#62 AndrejOrsula closed 3 years ago
1
Implement a custom replay buffer for octree observations

#61 AndrejOrsula closed 3 years ago
1
Add headless mode for ign_moveit (no RViz2)

#60 AndrejOrsula closed 3 years ago
0
Improve grasp detection for shaped reward function and add option for negative collision reward

#59 AndrejOrsula closed 3 years ago
0
Add Dockerfile and template for docker-compose

#58 AndrejOrsula closed 3 years ago
0
Use Fuel model of Panda by default

#57 AndrejOrsula closed 3 years ago
0
Merge hyperparameter train_freq with n_episodes_rollout

#56 AndrejOrsula closed 3 years ago
0