Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Episode Creator Script for Habitat 2.0 #764

Merged
merged 139 commits into from
Dec 10, 2021
Merged
Show file tree
Hide file tree
Changes from 136 commits
Commits
Show all changes
139 commits
Select commit Hold shift + click to select a range
0c4aa4d
Made episode ending more flexible
Aug 11, 2021
9be1277
Fixed typo
Aug 17, 2021
f9c395d
Temporary debugging
Aug 18, 2021
cbe782c
Added transform utils
Aug 18, 2021
14ad660
Working on EE tracking
Aug 20, 2021
5e2e1eb
Working on EE tracking
Aug 20, 2021
207a8d9
Added URDF config for IL
Aug 22, 2021
fb47b2e
First draft RearrangeEpisodeGenerator
aclegg3 Aug 27, 2021
77cfe00
CompositeArticulatedObjectSampler w/ rejection. DebugVisualizer objec…
aclegg3 Aug 27, 2021
4e78d2e
Updated reach settings
Aug 29, 2021
86e3182
Cleaned back up interactive play. Added oracle grasp
Aug 29, 2021
1c4669f
Fixed bug
Aug 30, 2021
53a52d9
Support sampling from AO receptacles. Simplify and improve sampling t…
aclegg3 Aug 30, 2021
1e2c369
omnidirectional peek util.
aclegg3 Aug 30, 2021
ed0db7a
cleanup
aclegg3 Aug 30, 2021
0cfea36
Updated Pick settings
Aug 31, 2021
32ee61a
cleanup and serialization test
aclegg3 Aug 31, 2021
b475a84
Fixed to rest metric
Aug 31, 2021
c264f48
add main to make rearrange_generator.py excecutable + parser args. Ad…
aclegg3 Aug 31, 2021
78b2f49
test serialization for generated dataset
aclegg3 Sep 1, 2021
2bcda49
cleanup
aclegg3 Sep 1, 2021
28de1f1
add receptacle name substring constraints for defining receptacle set…
aclegg3 Sep 1, 2021
ecb7466
minor changes and composite sampling example
aclegg3 Sep 1, 2021
4d8eb7f
Tilted Receptacle culling and definition of Receptacle local "up"
aclegg3 Sep 1, 2021
2945206
minor clean-up and rename expected tag bb_ to receptacle_
aclegg3 Sep 2, 2021
254ae05
fixes and counter receptacle definitions
aclegg3 Sep 3, 2021
e630192
refactor scene,object,receptacle set config format to fit yaml/yacs p…
aclegg3 Sep 13, 2021
083db37
migrate markers to yaml
aclegg3 Sep 13, 2021
092fbf5
refactor to support set exclusion and yaml compatible scene sampler c…
aclegg3 Sep 14, 2021
22044cc
migrate ao sampler to yaml friendly format
aclegg3 Sep 14, 2021
dba1404
migrate object and target samplers to yaml friendly format
aclegg3 Sep 14, 2021
af35e76
Merge remote-tracking branch 'origin/master' into rearrange-ep-generator
aclegg3 Sep 23, 2021
2db5d6c
Added sparse reward for reaching task
Sep 25, 2021
ce5d5ed
refactor to support deseralization of RearrangeEpisodes in new format
aclegg3 Sep 27, 2021
f59be28
adapt rearrange task configs to reference new dataset
aclegg3 Sep 28, 2021
0bb5a00
clean-up
aclegg3 Sep 28, 2021
15b0a13
clean-up
aclegg3 Sep 28, 2021
bbb2f8d
reformating for flake
aclegg3 Sep 28, 2021
b3655a3
Merge branch 'master' of https://github.com/facebookresearch/habitat-…
Oct 2, 2021
c08e20a
Added Rearrange Composite Task that can describe multistage rearrangm…
mathfac Sep 14, 2021
ec83979
Merge branch 'rearrange-ep-generator' of https://github.com/facebookr…
Oct 2, 2021
4a44d9c
Working on composite tasks
Oct 3, 2021
d94ccc8
Navigate to object navigation somewhat working
Oct 6, 2021
9418b94
Added benchmark
Oct 7, 2021
c1bb80b
Passing pre commit
Oct 7, 2021
943e296
Allowed more play options
Oct 12, 2021
5d502a3
Suction gripper, composite task
Oct 13, 2021
9286e0e
All tasks working smoothly
Oct 14, 2021
be201d8
Added concur render back
Oct 14, 2021
50101d7
Fixed rigid body removing bug. Added progress bar to episode generator
Oct 14, 2021
d28e4a4
Fixed up gym wrapper
Oct 14, 2021
18accb8
Refactored suction grip and added more tests
Oct 17, 2021
e2bda7c
Added star import
Oct 17, 2021
e04f04a
Added force terminate sensor
Oct 18, 2021
4d383c9
Removed prints
Oct 18, 2021
ab378df
auto-sleep and friends
aclegg3 Oct 18, 2021
1de2e06
Add benchmarking generator config
aclegg3 Oct 19, 2021
1479b63
Fixed bug, added episode init generation script
Oct 19, 2021
124614f
Merge branch 'hab_suite' of github.com:facebookresearch/habitat-lab i…
Oct 19, 2021
6847b93
update bnehcmark render option and bench scene
aclegg3 Oct 19, 2021
d04b130
Merge branch 'hab_suite' of https://github.com/facebookresearch/habit…
aclegg3 Oct 19, 2021
aa514e0
Added configurable robot type
Oct 19, 2021
6fc403e
Merge branch 'hab_suite' of github.com:facebookresearch/habitat-lab i…
Oct 19, 2021
568d165
Fixed the suction action
Oct 19, 2021
c39ed22
Fixed bad references
Oct 20, 2021
bff1c0e
update benchmark config to use no base Fetch
aclegg3 Oct 20, 2021
a9da80d
Fixed start position for idle benchmark
aclegg3 Oct 20, 2021
5ca5b50
update bench config to new bench scene
aclegg3 Oct 20, 2021
f53eaa8
Fixed final issue in non-existent constraints
Oct 21, 2021
6af76b0
Merge branch 'hab_suite' of github.com:facebookresearch/habitat-lab i…
Oct 21, 2021
aa39ed7
Added object sampling region scaling
Oct 22, 2021
746b8eb
Added convenience arguement to run the benchmark multiple times and a…
aclegg3 Oct 25, 2021
00256e1
Setting articulated object states and not resetting robot every time
Oct 25, 2021
62f9d7c
Merge branch 'hab_suite' of github.com:facebookresearch/habitat-lab i…
Oct 25, 2021
21b0c45
adjust rotation to point robot at the scene
aclegg3 Oct 25, 2021
ce359ca
Merge branch 'hab_suite' of https://github.com/facebookresearch/habit…
aclegg3 Oct 25, 2021
117f76b
Not-reloading objects between episode resets if object set is the same
Oct 26, 2021
7f0988d
Merge branch 'hab_suite' of github.com:facebookresearch/habitat-lab i…
Oct 26, 2021
d7a05ef
Fixed issue with constraint on episode reset
Oct 27, 2021
677bb82
Added all drawer tasks
Oct 28, 2021
f59683b
Added articulated task sensors
Oct 28, 2021
b8aa7ea
Added more gym options. Fixed up reach task
Oct 31, 2021
06c6f9e
Removed no wheels agent
Oct 31, 2021
fa9a041
Updated easy reach
Oct 31, 2021
7b046a3
Added any reach sensor to reach
Nov 1, 2021
545d5b2
add benchmark batch runner and plotting
aclegg3 Nov 3, 2021
95ee039
set bench debug video output to data/profile/
aclegg3 Nov 3, 2021
7098fca
enable optimization for concurrent rendering in benchmark config
aclegg3 Nov 3, 2021
b46ef92
add single camera idle benchmark config
aclegg3 Nov 3, 2021
ef78f54
add flag to skip robot update for benchmark
aclegg3 Nov 3, 2021
519e5a7
add back FetchRobotNoWheels
aclegg3 Nov 3, 2021
cd83230
minor bugfix - only update robot if physics stepping
aclegg3 Nov 4, 2021
7a1ad32
add types-pyyaml to support mypy pre-commit hook for yaml module useage
aclegg3 Nov 4, 2021
92174b0
python lint
aclegg3 Nov 4, 2021
3f74d69
update benchmark generator config for replicaCAD baked bench scene. A…
aclegg3 Nov 4, 2021
b17c805
Joint velocity sensor
Nov 7, 2021
0b97de4
Merge branch 'hab_suite' of github.com:facebookresearch/habitat-lab i…
Nov 7, 2021
8146615
bench changes to stepping and episode config
aclegg3 Nov 11, 2021
5312a09
Added many tests
Nov 15, 2021
f76e591
Merge branch 'hab_suite' of github.com:facebookresearch/habitat-lab i…
Nov 15, 2021
a08a61e
Switched stepping back to original
Nov 15, 2021
d0602d0
Updated configs
Nov 15, 2021
f0ca91e
changed action space to use ActionSpace class for DDPPO RL use
Nov 16, 2021
b7fd7b6
changed sensor observations for those that only output one 3D location
Nov 16, 2021
0755197
added max filename length for making videos
Nov 16, 2021
52c2477
added blacklist observations for viz-only images, automatic conversio…
Nov 16, 2021
f28268d
Merge branch 'hab_suite' of https://github.com/facebookresearch/habit…
Nov 16, 2021
a60dfda
Now using regular HabSim dataset source
Nov 16, 2021
f38de79
linting fixes
Nov 16, 2021
954df7a
added gaussian action space for rl_pick DDPPO training
Nov 16, 2021
9bb7429
Merge branch 'hab_suite' of github.com:facebookresearch/habitat-lab i…
Nov 16, 2021
36686d3
Add global receptacle handling for receptacles defined in the stage c…
aclegg3 Nov 18, 2021
26cc16f
Reverted action space
Nov 19, 2021
838ad98
Keeping only the episode creation code
Nov 24, 2021
2c9d77b
Tests should all pass now
Nov 24, 2021
dd95885
Added license
Nov 24, 2021
b6cbea0
Fixed MP test
Nov 24, 2021
f685390
Fixed example
Nov 25, 2021
bf43de3
Addressed Aarons comments
Nov 27, 2021
d8be611
Added desired resting arm end-effector position to config
Nov 27, 2021
6683c91
Cleaned up code
Nov 29, 2021
2a78fbf
Fixed recording and playing back trajectories in play script
Nov 30, 2021
8cbfedd
removed unused obj_loaders.py utils and refactored code to remove dep…
aclegg3 Dec 2, 2021
555825c
removed leftover function call
aclegg3 Dec 2, 2021
e022a67
address reviews: clean-up some docstrings, separate receptacle file, …
aclegg3 Dec 2, 2021
71492be
Removing outdated Habitat_Interactive_Tasks tutorial colab for now. W…
aclegg3 Dec 2, 2021
76b47c2
remove test reference to deleted tutorial
aclegg3 Dec 2, 2021
6b39b19
Receptacle refactor. Split into abstract Recetpacle and derived AABBR…
aclegg3 Dec 3, 2021
4dd4dee
bugfix: fill env config values from current episode instead of first
aclegg3 Dec 3, 2021
1085d57
address minor bug and review comments
aclegg3 Dec 3, 2021
d0ecd2c
Add/update documentation
aclegg3 Dec 3, 2021
038f2fd
refactor to move generic utilities out of habitat/datasets/rearrange
aclegg3 Dec 3, 2021
7febbc2
rearrange_sim.py docstring and typing
aclegg3 Dec 8, 2021
d9521dd
Add some docs for RearrangeGenerator
aclegg3 Dec 8, 2021
77681b4
Add some docstrings for actions.py
aclegg3 Dec 8, 2021
97153cb
rearrange_graps_manager.py - docstrings and typing
aclegg3 Dec 8, 2021
598b6ef
fix circular import issues
aclegg3 Dec 8, 2021
7f9cbce
Merge remote-tracking branch 'origin/main' into episode_creator
aclegg3 Dec 9, 2021
4614fbd
Improved IK warning message
Dec 10, 2021
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -107,14 +107,14 @@ If you use the Habitat platform in your research, please cite the [Habitat](http
## Example
<!--- Please, update `examples/example.py` if you update example. -->

🆕Example code-snippet which uses [`tasks/rearrangepick_replica_cad.yaml`](configs/tasks/rearrangepick_replica_cad.yaml) for configuration of task and agent.
🆕Example code-snippet which uses [`configs/tasks/rearrange/pick.yaml`](configs/tasks/rearrange/pick.yaml) for configuration of task and agent.

```python
import habitat

# Load embodied AI task (RearrangePick) and a pre-specified virtual robot
env = habitat.Env(
config=habitat.get_config("configs/tasks/rearrangepick_replica_cad.yaml")
config=habitat.get_config("configs/tasks/rearrange/pick.yaml")
)

observations = env.reset()
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -3,17 +3,24 @@ ENVIRONMENT:
DATASET:
TYPE: RearrangeDataset-v0
SPLIT: train
DATA_PATH: data/datasets/rearrange_pick/replica_cad/v0/{split}/{split}_counter_L_analysis_5000_500.json.gz
DATA_PATH: data/datasets/rearrange_pick/replica_cad/v0/rearrange_pick_replica_cad_v0/pick.json.gz
SCENES_DIR: "data/replica_cad/"
TASK:
TYPE: RearrangePickTask-v0
MAX_COLLISIONS: -1.0
COUNT_OBJ_COLLISIONS: True
COUNT_ROBOT_OBJ_COLLS: False
DESIRED_RESTING_POSITION: [0.5, 0.0, 1.0]

# In radians
BASE_ANGLE_NOISE: 0.15
BASE_NOISE: 0.05
CONSTRAINT_VIOLATE_SHOULD_END: True
ASzot marked this conversation as resolved.
Show resolved Hide resolved
FORCE_REGENERATE: False

# Measurements for composite tasks.
REWARD_MEASUREMENT: "rearrangepick_reward"
SUCCESS_MEASUREMENT: "rearrangepick_success"

# If true, does not care about navigability or collisions with objects when spawning
# robot
Expand Down Expand Up @@ -44,52 +51,52 @@ TASK:
TYPE: "IsHoldingSensor"
RELATIVE_RESTING_POS_SENSOR:
TYPE: "RelativeRestingPositionSensor"
LOCAL_SENSOR:
TYPE: "LocalizationSensor"
SENSORS: ["TARGET_START_SENSOR", "GOAL_SENSOR", "JOINT_SENSOR", "IS_HOLDING_SENSOR", "END_EFFECTOR_SENSOR", "LOCAL_SENSOR", "RELATIVE_RESTING_POS_SENSOR"]
SENSORS: ["TARGET_START_SENSOR", "JOINT_SENSOR", "IS_HOLDING_SENSOR", "END_EFFECTOR_SENSOR", "RELATIVE_RESTING_POS_SENSOR"]
ROBOT_FORCE:
TYPE: "RobotForce"
MIN_FORCE: 20.0
FORCE_TERMINATE:
ASzot marked this conversation as resolved.
Show resolved Hide resolved
TYPE: "ForceTerminate"
MAX_ACCUM_FORCE: 5000.0
ROBOT_COLLS:
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Let's be more verbose here:

Suggested change
ROBOT_COLLS:
ROBOT_COLLISIONS:

TYPE: "RobotCollisions"
TYPE: "RobotCollisions"
OBJECT_TO_GOAL_DISTANCE:
TYPE: "ObjectToGoalDistance"
END_EFFECTOR_TO_OBJECT_DISTANCE:
TYPE: "EndEffectorToObjectDistance"
END_EFFECTOR_TO_REST_DISTANCE:
TYPE: "EndEffectorToRestDistance"
REARRANGE_PICK_REWARD:
PICK_REWARD:
TYPE: "RearrangePickReward"
DIST_REWARD: 20.0
SUCC_REWARD: 10.0
PICK_REWARD: 20.0
CONSTRAINT_VIOLATE_PEN: 10.0
DROP_PEN: 5.0
WRONG_PICK_PEN: 5.0
COLL_PEN: 1.0
ROBOT_OBJ_COLL_PEN: 0.0
MAX_ACCUM_FORCE: 5000.0
USE_DIFF: True
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Need more explicit name here, not clear what "use" means.

DROP_OBJ_SHOULD_END: False
WRONG_PICK_SHOULD_END: False

# General Rearrange Reward config
CONSTRAINT_VIOLATE_PEN: 10.0
FORCE_PEN: 0.001
MAX_FORCE_PEN: 1.0
FORCE_END_PEN: 10.0
USE_DIFF: True

DROP_OBJ_SHOULD_END: False
WRONG_PICK_SHOULD_END: False

REARRANGE_PICK_SUCCESS:
PICK_SUCCESS:
TYPE: "RearrangePickSuccess"
SUCC_THRESH: 0.15
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Naming similar to GRASP_THRESH_DIST :

Suggested change
SUCC_THRESH: 0.15
SUCC_THRESH_DIST: 0.15


MEASUREMENTS:
- "OBJECT_TO_GOAL_DISTANCE"
- "ROBOT_FORCE"
- "FORCE_TERMINATE"
- "ROBOT_COLLS"
- "END_EFFECTOR_TO_REST_DISTANCE"
- "END_EFFECTOR_TO_OBJECT_DISTANCE"
- "REARRANGE_PICK_SUCCESS"
- "REARRANGE_PICK_REWARD"
- "PICK_SUCCESS"
- "PICK_REWARD"
ACTIONS:
ARM_ACTION:
TYPE: "ArmAction"
Expand All @@ -99,12 +106,14 @@ TASK:
GRASP_THRESH_DIST: 0.15
DISABLE_GRIP: False
DELTA_POS_LIMIT: 0.0125
EE_CTRL_LIM: 0.015
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Let's use naming similar to DELTA_POS_LIMIT:

Suggested change
EE_CTRL_LIM: 0.015
EE_CTRL_LIMIT: 0.015

POSSIBLE_ACTIONS:
- ARM_ACTION

SIMULATOR:
ACTION_SPACE_CONFIG: v0
AGENTS: ['AGENT_0']
ROBOT_JOINT_START_NOISE: 0.0
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

A comment needed for unit of measurement of the noise.

AGENT_0:
HEIGHT: 1.5
IS_SET_START_STATE: False
Expand Down Expand Up @@ -136,8 +145,8 @@ SIMULATOR:
CTRL_FREQ: 120.0
AC_FREQ_RATIO: 4
ROBOT_URDF: ./data/robots/hab_fetch/robots/hab_fetch.urdf
ROBOT_TYPE: "FetchRobot"
FORWARD_STEP_SIZE: 0.25
LAG_OBSERVATIONS: 1

# Grasping
HOLD_THRESH: 0.09
Expand Down
Original file line number Diff line number Diff line change
@@ -1,21 +1,23 @@
# The configuration to run the SensePlanAct baseline architecture from https://arxiv.org/abs/2106.14405
ENVIRONMENT:
MAX_EPISODE_STEPS: 200
MAX_EPISODE_STEPS: 30
DATASET:
TYPE: RearrangeDataset-v0
SPLIT: train
DATA_PATH: data/datasets/rearrange_pick/replica_cad/v0/{split}/{split}_counter_L_analysis_5000_500.json.gz
DATA_PATH: data/datasets/rearrange_pick/replica_cad/v0/rearrange_pick_replica_cad_v0/pick.json.gz
SCENES_DIR: "data/replica_cad/"
TASK:
TYPE: RearrangePickTask-v0
MAX_COLLISIONS: -1.0
COUNT_OBJ_COLLISIONS: True
COUNT_ROBOT_OBJ_COLLS: False
CONSTRAINT_VIOLATE_SHOULD_END: True

# In radians
BASE_ANGLE_NOISE: 0.15
BASE_NOISE: 0.05
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

A comment needed for unit of measurement of the noise.

FORCE_RECACHE: False
FORCE_REGENERATE: False
DESIRED_RESTING_POSITION: [0.5, 0.0, 1.0]

# If true, does not care about navigability or collisions with objects when spawning
# robot
Expand Down Expand Up @@ -48,11 +50,9 @@ TASK:
TYPE: "EEPositionSensor"
IS_HOLDING_SENSOR:
TYPE: "IsHoldingSensor"
LOCAL_SENSOR:
TYPE: "LocalizationSensor"
RELATIVE_RESTING_POS_SENSOR:
TYPE: "RelativeRestingPositionSensor"
SENSORS: ["TARGET_START_SENSOR", "GOAL_SENSOR", "JOINT_SENSOR", "IS_HOLDING_SENSOR", "END_EFFECTOR_SENSOR", "LOCAL_SENSOR", "TARGET_CUR_SENSOR", "TARGET_CUR_SENSOR", "RELATIVE_RESTING_POS_SENSOR"]
SENSORS: ["TARGET_START_SENSOR", "GOAL_SENSOR", "JOINT_SENSOR", "IS_HOLDING_SENSOR", "END_EFFECTOR_SENSOR", "TARGET_CUR_SENSOR", "TARGET_CUR_SENSOR", "RELATIVE_RESTING_POS_SENSOR"]
ROBOT_FORCE:
TYPE: "RobotForce"
MIN_FORCE: 20.0
Expand All @@ -64,37 +64,37 @@ TASK:
TYPE: "EndEffectorToObjectDistance"
END_EFFECTOR_TO_REST_DISTANCE:
TYPE: "EndEffectorToRestDistance"
REARRANGE_PICK_REWARD:
FORCE_TERMINATE:
TYPE: "ForceTerminate"
MAX_ACCUM_FORCE: 0.0
PICK_REWARD:
TYPE: "RearrangePickReward"
DIST_REWARD: 20.0
SUCC_REWARD: 10.0
PICK_REWARD: 20.0
DIST_REWARD: 10.0
PICK_REWARD: 5.0
CONSTRAINT_VIOLATE_PEN: 10.0
DROP_PEN: 5.0
WRONG_PICK_PEN: 5.0
COLL_PEN: 1.0
COLL_PEN: 0.0
ROBOT_OBJ_COLL_PEN: 0.0
MAX_ACCUM_FORCE: null

FORCE_PEN: 0.001
MAX_FORCE_PEN: 1.0
FORCE_END_PEN: 10.0
FORCE_PEN: 0.0
MAX_FORCE_PEN: 0.0
FORCE_END_PEN: 0.0
USE_DIFF: True

DROP_OBJ_SHOULD_END: False
WRONG_PICK_SHOULD_END: False

REARRANGE_PICK_SUCCESS:
PICK_SUCCESS:
TYPE: "RearrangePickSuccess"
SUCC_THRESH: 0.3

SUCC_THRESH: 0.15
MEASUREMENTS:
- "ROBOT_FORCE"
- "ROBOT_COLLS"
- "END_EFFECTOR_TO_OBJECT_DISTANCE"
- "FORCE_TERMINATE"
- "END_EFFECTOR_TO_REST_DISTANCE"
- "REARRANGE_PICK_SUCCESS"
- "REARRANGE_PICK_REWARD"
- "END_EFFECTOR_TO_OBJECT_DISTANCE"
- "PICK_SUCCESS"
- "PICK_REWARD"
ACTIONS:
ARM_ACTION:
TYPE: "ArmAction"
Expand All @@ -105,6 +105,7 @@ TASK:
DISABLE_GRIP: False
DELTA_POS_LIMIT: 0.0125
EE_CTRL_LIM: 0.015
ORACLE_GRASP: True
EMPTY:
TYPE: "EmptyAction"
POSSIBLE_ACTIONS:
Expand All @@ -114,6 +115,7 @@ TASK:
SIMULATOR:
ACTION_SPACE_CONFIG: v0
AGENTS: ['AGENT_0']
ROBOT_JOINT_START_NOISE: 0.0
AGENT_0:
HEIGHT: 1.5
IS_SET_START_STATE: False
Expand Down Expand Up @@ -146,11 +148,11 @@ SIMULATOR:
# Agent setup
ARM_REST: [0.6, 0.0, 0.9]
CTRL_FREQ: 120.0
AC_FREQ_RATIO: 4
AC_FREQ_RATIO: 8
ROBOT_URDF: ./data/robots/hab_fetch/robots/hab_fetch.urdf
ROBOT_TYPE: "FetchRobot"
IK_ARM_URDF: ./data/robots/hab_fetch/robots/fetch_onlyarm.urdf
FORWARD_STEP_SIZE: 0.25
LAG_OBSERVATIONS: 1

# Grasping
HOLD_THRESH: 0.09
Expand Down
Loading