Release Notes
OMY
https://github.com/ROBOTIS-GIT/open_manipulator
4.0.5 (2025-07-02)
- Added init_position_file argument to launch files
- Updated profile time and acceleration time for OMY series
- Fixed lint errors
- Updated Collision area for OMY series
- Added feedback control for leader-follower synchronization
- Contributors: Woojin Wie, Sungho Woo
Physical AI Tools
0.5.9 (2025-07-07)
- Use global ROS connection instead of multiple instances
- Add proper cleanup for image streams to prevent accumulation
- Remove unnecessary scrollbars in Chrome browser
- Contributors: Kiwoong Park
0.5.8 (2025-07-07)
- Applied Redux Toolkit for better state management
- Added heartbeat status to the UI
- Added heartbeat topic publishing to monitor alive status of Physical AI Server
- Contributors: Kiwoong Park, Dongyun Kim
0.5.7 (2025-06-26)
- Added Image Transport Plugin and fixed missing Gstreamer components
- Contributors: Dongyun Kim
0.5.6 (2025-06-26)
- Reordered pip install order in Dockerfile to fix the numpy version issue
- Contributors: Woojin Wie
0.5.5 (2025-06-26)
- Fixed control panel button states not reflecting correct taskType when switching between Record and Inference pages
- Contributors: Kiwoong Park
0.5.4 (2025-06-25)
- Added support for inference mode in the physical AI Server, including a new InferencePage and related UI components.
- Changed the robot naming format.
- Added Robot Config to support FFW-SG2 robot.
- Added Msg Topic and data acquisition functionality to support Mobile Robot.
- Fixed minor errors in the data acquisition process to improve software stability.
- Added a new inference page for running and monitoring inference tasks.
- Added inference-related msgs and srv types.
- Contributors: Dongyun Kim, Kiwoong Park
0.5.3 (2025-06-16)
- Refactored Physical AI Server for improved data collection capabilities
- Implemented data acquisition functionality using ROS2 topics
- Modified configuration system to allow flexible robot type selection
- Updated data collection method to utilize image buffers for efficiency
- Overall UI improvements for physical_ai_manager
- Added status information display from physical_ai_server
- Added functionality to receive task information from users and send commands to physical_ai_server
- Added bringup launch file that runs physical_ai_server with rosbridge_server and webvideo_server
- Contributors: Dongyun Kim, Kiwoong Park
0.5.2 (2025-05-29)
- Adjusted the waiting timeout for joint states.
- Contributors: Dongyun Kim
0.5.1 (2025-05-29)
- Added quality and transport parameters to image streaming URL
- Added a Docker setup for physical AI server
- Contributors: Kiwoong Park
0.5.0 (2025-05-20)
- Added a web UI for physical AI data collection
- Removed unnecessary dependencies and cleaned up the codebase
- Updated the LeRobot submodule to the latest version
- Refactored to a scalable structure that supports N cameras and various joint configurations
- Contributors: Dongyun Kim, Kiwoong Park, Woojin Wie, Seongwoo Kim
0.4.0 (2025-05-15)
- Added a pipeline for data collection and inference based on ROS2
- Refactored to a scalable structure that supports N cameras and various joint configurations
- Contributors: Dongyun Kim
0.3.1 (2025-05-08)
- Updated the LeRobot submodule to the latest version
- Contributors: Woojin Wie
0.3.0 (2025-04-25)
- Unified multiple launch files into a single configurable launch file for better usability
- Contributors: Seongwoo Kim
0.2.0 (2025-04-08)
- Added a time stamper node for data synchronization purposes
- Removed unused joints and motors bus config
- Contributors: Seongwoo Kim, Hyungyu Kim
0.1.0 (2025-04-07)
- Added a full workflow for recording and visualizing datasets using the LeRobot interface
- Added bringup scripts for system initialization
- Contributors: Seongwoo Kim, Pyo