Semi-Autonomous Teleoperation via Learning Non-Prehensile Manipulation Skills
This work addresses the challenge of efficient object manipulation in cluttered settings for robotics and teleoperation applications, representing an incremental improvement over existing methods.
The paper tackles the problem of semi-autonomous teleoperation for pick-and-place tasks in cluttered environments by learning non-prehensile manipulation skills via reinforcement learning, resulting in a method that outperforms manual keyboard control by reducing the time duration for grasping in real-world experiments.
In this paper, we present a semi-autonomous teleoperation framework for a pick-and-place task using an RGB-D sensor. In particular, we assume that the target object is located in a cluttered environment where both prehensile grasping and non-prehensile manipulation are combined for efficient teleoperation. A trajectory-based reinforcement learning is utilized for learning the non-prehensile manipulation to rearrange the objects for enabling direct grasping. From the depth image of the cluttered environment and the location of the goal object, the learned policy can provide multiple options of non-prehensile manipulation to the human operator. We carefully design a reward function for the rearranging task where the policy is trained in a simulational environment. Then, the trained policy is transferred to a real-world and evaluated in a number of real-world experiments with the varying number of objects where we show that the proposed method outperforms manual keyboard control in terms of the time duration for the grasping.