BCN3D Moveo 커뮤니티 빌드
교육용 오픈소스 3D 프린팅 5축 로봇 암
라이선스
MIT License
- Complete ROS + MoveIt + TensorFlow Control Platform for BCN3D Moveo (Jesse Weisberg): A complete ROS control platform for the BCN3D Moveo open source robot arm, featuring MoveIt motion planning, Gazebo simulation, and TensorFlow-based object detection for autonomous pick-and-place tasks. What Was Built Jesse Weisberg took the BCN3D Moveo — an open source 3D-printed 5-axis robot arm — and developed a comprehensive software stack that transforms it from a mechanical project into a fully programmable robotics platform: - Full ROS integration with custom URDF model, joint state publishers, and hardware interface via Arduino + rosserial - MoveIt motion planning for collision-aware path planning with RViz visualization - Gazebo simulation environment for testing movements before running on hardware - TensorFlow object detection using SSD MobileNet for real-time object recognition - Autonomous pick-and-place pipeline combining vision + motion planning Technical Architecture The system uses an Arduino Mega 2560 with RAMPS 1.4 stepper driver shield as the low-level controller, communicating with ROS via rosserial. The ROS node graph handles: 1. Joint trajectory control — MoveIt generates collision-free trajectories, which are sent to Arduino 2. Inverse kinematics — Custom IK solver maps end-effector poses to joint angles 3. Visual perception — TensorFlow model identifies objects in webcam feed, localizes them in 3D space 4. Task execution — Coordinates perception and planning for pick-and-place sequences Why It Matters With 322+ GitHub stars and an MIT license, this project became one of the most popular ROS packages for affordable robot arms. It demonstrated that a sub-$400 3D-printed arm could run the same software stack (ROS + MoveIt + perception) used by industrial robots costing 100x more. The project is frequently cited in robotics education and has been replicated by university labs and hobbyists worldwide. Jesse's detailed project page walks through the entire build process.