REPLAB
The project’s goal is to replicate a standard, reproducible, and cheap environment for benchmarking robotic arm object grasping algorithms. The original environment was created in Berkley and reviewed in the article:
“REPLAB: A Reproducible Low-Cost Arm Benchmark Platform for
Robotic Learning” by Brian Yang, Jesse Zhang, Vitchyr Pong, Sergey Levine, and Dinesh Jayaraman.
The standard REPLAB cell includes an arena with a wooden base, a cage comprised of metal beams, WIDOWX MK II robotic arm of TrossenRobotics, and a 3D camera. We used as a 3D camera the Intel RealSense D435. This is the only difference from the original environment.
After cell Construction and integration with the D435 camera, Calibrations were held and followed by the evaluation phase of the cell’s performance.
During the evaluation process grasp data was collected and labeled to allow further training and research using the platform.