SOURCE-LINKED INTELLIGENCE

SE(3) Neural Potential Fields for 6-DoF Trajectory Planning Directly from Images Without Explicit 3D Reconstruction

arXiv · AI, language, vision and robotics · article · Sep 21, 2026 · UTC

Reaching a 6-DoF grasp pose in clutter requires a collision-free trajectory, conventionally obtained by reconstructing the scene in 3D and planning inside that reconstruction, at the cost of its accuracy and compute. Potential fields learned directly from images remove that dependency but inherit the classical weakness of artificial potential fields: where attractive and repulsive gradients cancel, the descent grazes the obstacle instead of going around it, and can stall short of the goal. We present an SE(3) neural potential field learned from posed RGB images and supervised with a navigation

Read original source ↗ Open in workspace

recordType

paper

region

Global