YAM Cell

✓ Python SDK✓ Bimanual✓ Data Generation

YAM Cell is a complete bimanual teleoperation workstation built around two YAM leader arms and two YAM follower arms. It is designed for collecting high-quality manipulation demonstrations for training embodied AI models.

Mobile Variant

System Overview

The YAM Cell pairs leader arms (with teaching handles) and follower arms in a bilateral teleoperation loop. An operator moves the leader arms naturally; the follower arms mirror the motion in real time with configurable force feedback.

Operator ──► Leader (YAM + teaching handle)
               │  bilateral link (CAN bus)
             Follower (YAM + task gripper)  ──► Task workspace

Hardware Requirements

Component	Qty	Notes
YAM Follower arms	2	Any YAM tier
YAM Leader arms	2	Must use yam_teaching_handle gripper
CANable USB-CAN adapters	4	One per arm
Workstation PC	1	Ubuntu 22.04 recommended

CAN Bus Layout

Each arm requires a dedicated CAN channel. Assign persistent names using udev rules (see Hardware Setup):

Arm	CAN name
Left follower	can_follower_l
Right follower	can_follower_r
Left leader	can_leader_l
Right leader	can_leader_r

Videos

🎬

Video — Coming Soon

Operator sitting in front of two YAM leader arms, controlling two follower arms picking objects. Side-by-side view of leader and follower workspace. 1–2 min demo.

Quick Start

1. Configure CAN IDs

Plug one CANable device at a time and assign each arm a persistent name:

# Follow the instructions in:
doc/set_persist_id_socket_can.md

2. Verify connectivity

ip a | grep can
# Expected:
# can_follower_r  UP
# can_follower_l  UP
# can_leader_r    UP
# can_leader_l    UP

3. Launch bimanual teleoperation

cd /path/to/i2rt
source .venv/bin/activate
python examples/bimanual_lead_follower/bimanual_lead_follower.py

4. Enable synchronization

Press the top button on either teaching handle once to begin tracking. Press again to pause.

Control Reference

Action	Result
Move leader arm	Follower mirrors motion
Squeeze trigger	Follower gripper closes
Top button (1×)	Enable sync
Top button (2×)	Pause sync
--bilateral_kp flag	Increase for more force feedback (default 0.2)

Bilateral stiffness

Start with --bilateral_kp 0.1 and increase gradually. Values above 0.3 make the leader arm feel noticeably heavy.

Data Logging

Trajectory recording is built into the example:

from i2rt.robots.motor_chain_robot import get_yam_robot
# See examples/record_replay_trajectory/ for full dataset collection pipeline

See Also

• Bimanual Teleoperation Example
• Record & Replay
• YAM Arm API