This was a very interesting talk. Thanks for contributing to the research community by opening such a model and providing all the required details and insights!
Can OpenVLA handle change in the action space for example, instead of grabbing with 2 fingers, if I want the robot to grasp the object based on a suction-based gripper?
This was a very interesting talk. Thanks for contributing to the research community by opening such a model and providing all the required details and insights!
Very interesting work, where is the list of robots used for the first dataset/model?
So how can I transform the output 7 action figures into the real length?
Can OpenVLA handle change in the action space for example, instead of grabbing with 2 fingers, if I want the robot to grasp the object based on a suction-based gripper?