Denoising Heat-inspired Diffusion with Insulators for Collision Free Motion Planning
Junwoo Chang*, Hyunwoo Ryu*, Jiwoo Kim, Soochul Yoo, Jongeun Choi, Joohwan Seo, Nikhil Prakash, Roberto Horowitz
(* : equally contributed)
Neurips 2023 Workshop on Diffusion Models
Abstract
Diffusion models have risen as a powerful tool in robotics due to their flexibility and multi-modality. While some of these methods effectively address complex problems, they often depend heavily on inference-time obstacle detection and require additional equipment. Addressing these challenges, we present a method that, during inference time, simultaneously generates only reachable goals and plans motions that avoid obstacles, all from a single visual input. Central to our approach is the novel use of a collision-avoiding diffusion kernel for training. Through evaluations against behavior-cloning and classical diffusion models, our framework has proven its robustness. It is particularly effective in multi-modal environments, navigating toward goals and avoiding unreachable ones blocked by obstacles, while ensuring collision avoidance.
Demos
All experiments are conducted to plan the sample positions of the agent (red dots) toward the goals (apples) while avoiding obstacles (black rectangles).
U map (Experiment) :
Multiple goals
(all reachable)
Multiple goals
(one unreachable)
When a collision occurs, the red dots get stuck in the obstacles, indicating that if no stationary dots remain, no collisions are occurring.
Other maps :
Multiple goals
(all reachable)
Multiple goals
(one unreachable)
Demo with heat-inspired diffusion kernel
