google-deepmind / google-deepmind/dm_control
How can I get concrete definitions for the results from `physics.named.data`
- Dominant language
- Python
- Stars
- 4.7k
- Forks
- 764
- PR merge metrics
- No merged PRs in 30d
Description
Take the "quadruped" domain as an example. It is known that we can get rough understandings from the `quadruped.py` that `self.named.data.sensordata` contains sensor readings as follows:
```
self.named.data.sensordata
Out[33]:
FieldIndexer(sensordata):
0 imu_accel [ 7.2 ]
1 imu_accel [-2.65 ]
2 imu_accel [ 22.7 ]
3 imu_gyro [ 0.448 ]
4 imu_gyro [ 0.411 ]
5 imu_gyro [-0.751 ]
6 velocimeter [-0.206 ]
7 velocimeter [-0.0947 ]
8 velocimeter [-0.249 ]
9 force_toe_front_left [-62.4 ]
10 force_toe_front_left [ 30.4 ]
11 force_toe_front_left [-1.41 ]
12 force_toe_front_right [ 85.1 ]
13 force_toe_front_right [ 20.3 ]
14 force_toe_front_right [-101 ]
15 force_toe_back_right [ 1.19e+03]
16 force_toe_back_right [-291 ]
17 force_toe_back_right [-725 ]
18 force_toe_back_left [ 256 ]
19 force_toe_back_left [-23.1 ]
20 force_toe_back_left [-936 ]
21 torque_toe_front_left [ 0.232 ]
22 torque_toe_front_left [ 0.757 ]
23 torque_toe_front_left [-0.187 ]
24 torque_toe_front_right [-0.256 ]
25 torque_toe_front_right [-1.01 ]
26 torque_toe_front_right [ 0.139 ]
27 torque_toe_back_right [ 2.94 ]
28 torque_toe_back_right [-19.2 ]
29 torque_toe_back_right [ 13 ]
30 torque_toe_back_left [-28.8 ]
31 torque_toe_back_left [ 36.2 ]
32 torque_toe_back_left [-9.05 ]
33 center_of_mass [-0.00071 ]
34 center_of_mass [-0.0362 ]
35 center_of_mass [ 0.82 ]
```
Although these names provide basic information, I still don't know the concrete meanings for each item. For example, it is hard to infer which `imu_accel ` is for x-axis.
Contributor guide
Research direction
Start with quadruped.py and the self.named.data.sensordata output shown in the issue. Trace the sensor definitions to determine the axis and physical meaning of each grouped value, including imu_accel, imu_gyro, force, torque, and center_of_mass entries. Done means documenting concrete meanings for the reported indices so users can interpret each result.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- documentation
- Issue type
- Documentation
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100