LeRobot Arm
Robotic education platform project:
| Aspect | Details |
|---|---|
| Control | Master-slave logic, ESP32 control |
| Application | Teaching, interactive demos |
Hardware Assembly & Linkage Verification
We have successfully assembled the LeRobot hardware and established the foundational control systems. Our current milestone focused on the Master-Slave Linkage control mechanism.
By manually manipulating the master arm, we continuously track its joint angles and transmit this data in real-time to the slave arm. The slave arm precisely mimics the master's movements, allowing for intuitive and fluid teleoperation.
We performed several grasping verification tests, demonstrating fine-grained control over the end-effector. The linkage proves to be highly responsive, making it an excellent platform for gathering imitation learning datasets.
Grasping Test
The arm dynamically adjusts to different shapes and weights during grasping tests.
Master-Slave Linkage
The primary controller arm (master) and the actuated arm (slave) mirror each other in real-time.
Teleoperation
Operators can easily manipulate the environment from a safe distance using the responsive master arm setup.
Future Roadmap: Vision & Autonomy
Moving forward, our next major objective is to integrate environmental perception and autonomous capabilities.
- Camera Integration: We plan to mount a camera on the robotic setup to capture real-time visual data of the workspace.
- Image Recognition: Utilizing computer vision models to identify and locate target objects dynamically.
- Autonomous Grasping: Transitioning from pure teleoperation to AI-driven autonomous grasping, where the robot can plan its own trajectories and manipulate objects based on visual input and learned policies.