Remote Control of Robotic Arms – Using a Standard Gamepad
Open-source project maps joysticks to joint control, enabling intuitive teleoperation of a 6-DOF PiPER manipulator.
Agilex Robotics, in collaboration with developer Kehuan Jack, has open-sourced 'Gamepad_PiPER,' a software package that enables intuitive teleoperation of their PiPER 6-degree-of-freedom robotic arm using any standard USB or Bluetooth gamepad, like those from Xbox or PlayStation. The project, hosted on GitHub, maps the familiar controls of a gamepad to complex robotic functions: the left joystick controls the base rotation and shoulder joint, the right joystick handles the elbow and wrist rotation, and triggers/buttons manage the gripper, speed, and pose saving. A key feature is its dual-mode operation, allowing users to switch between direct 'Joint Control' of individual actuators and higher-level 'Pose Control' for moving the arm's end-effector in Cartesian space, which is handled automatically by integrated inverse kinematics solvers.
The system is designed for the Robot Operating System (ROS) environment and offers four different backend kinematics libraries for flexibility, including Pinocchio, PyRoKi, cuRobo, and the recommended pytracik. This allows developers and researchers to choose the best solver for their needs, whether prioritizing speed, accuracy, or GPU acceleration. A live web visualization at http://localhost:8080 provides real-time feedback of the arm's position. The setup process is documented for Ubuntu 20.04+ with Python 3.9+, and the repository includes a virtual mode (`main_virtual.py`) for safe testing in simulation before deploying to physical hardware, lowering the barrier for experimentation and prototyping.
- Maps standard gamepad inputs to control all 6 joints, gripper, and end-effector pose of a PiPER robotic arm.
- Supports four different kinematics backends (Pinocchio, PyRoKi, cuRobo, pytracik) for inverse kinematics calculations.
- Includes a web-based visualizer and a simulation mode for safe testing before real-world deployment.
Why It Matters
Drastically lowers the skill barrier for robotic teleoperation, enabling faster prototyping, research, and educational demos with affordable, familiar hardware.