2.3 KiB
2.3 KiB
2D Walker Robot
The 2D Walker Robot (Walker2D) is a classic robot control task from DeepMind Control Suite. The goal is to achieve standing, walking, and running by controlling the robot's joints.
:poster: _static/images/poster/dm_walker.jpg
:nocontrols:
:autoplay:
:playsinline:
:muted:
:loop:
:width: 100%
Task Description
Walker2D is a 2D planar bipedal robot with multiple joints and actuators:
- State Space: Includes rotation angles and angular velocities of various robot parts, torso height and velocity, etc.
- Action Space: Control torques for each joint
- Reward Function: Mainly composed of maintaining standing balance and forward speed
- Termination Conditions: Robot falls or joints reach limit positions
Three Task Modes
- dm-stander: Static standing task (move_speed = 0.0)
uv run scripts/train.py --env dm-stander
- dm-walker: Walking task (move_speed = 1.0)
uv run scripts/train.py --env dm-walker
- dm-runner: Running task (move_speed = 5.0)
uv run scripts/train.py --env dm-runner
Quick Start
1. Environment Preview
uv run scripts/view.py --env dm-stander
uv run scripts/view.py --env dm-walker
uv run scripts/view.py --env dm-runner
2. Start Training
uv run scripts/train.py --env dm-stander
uv run scripts/train.py --env dm-walker
uv run scripts/train.py --env dm-runner
3. View Training Progress
uv run tensorboard --logdir runs/dm-walker
4. Test Training Results
uv run scripts/play.py --env dm-stander
uv run scripts/play.py --env dm-walker
uv run scripts/play.py --env dm-runner
Reward Function Design
Walker2D's reward function consists of the following components:
Basic Standing Reward
# Height reward: keep torso at target height
# Upright reward: keep torso upright
Movement Reward (walking and running tasks)
# Speed reward: track target speed
# Total reward = standing reward * movement weight
Expected Results
-
dm-stander:
- Torso height maintained in 1.0-1.4m range
-
dm-walker:
- Actual walking speed close to 1.0 m/s
-
dm-runner:
- Running speed reaches 4.0-5.0 m/s