Neural Motion Fields: Encoding Grasp Trajectories as Implicit Value Functions
This addresses a key challenge in robotics for dynamic object manipulation, representing a novel approach rather than an incremental improvement.
The paper tackles the problem of grasping dynamic objects in constrained environments by introducing Neural Motion Fields, an object representation that encodes point clouds and task trajectories as an implicit value function, enabling reactive grasping with sampling-based MPC.
The pipeline of current robotic pick-and-place methods typically consists of several stages: grasp pose detection, finding inverse kinematic solutions for the detected poses, planning a collision-free trajectory, and then executing the open-loop trajectory to the grasp pose with a low-level tracking controller. While these grasping methods have shown good performance on grasping static objects on a table-top, the problem of grasping dynamic objects in constrained environments remains an open problem. We present Neural Motion Fields, a novel object representation which encodes both object point clouds and the relative task trajectories as an implicit value function parameterized by a neural network. This object-centric representation models a continuous distribution over the SE(3) space and allows us to perform grasping reactively by leveraging sampling-based MPC to optimize this value function.