Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.21460 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913150170824704 |
|---|---|
| author | Zilka, Riley Khlynovskiy, Sergey Wang, Allie Jagersand, Martin |
| author_facet | Zilka, Riley Khlynovskiy, Sergey Wang, Allie Jagersand, Martin |
| contents | Autonomous manipulation systems have achieved remarkable capabilities, yet the integration of human expertise with diffusion-based policies in shared control remains relatively unexplored. In this paper, we propose Human-In-The-Loop Diffusion (HITL-D), a shared control framework that enhances user performance in multi-step, insertion, and fine manipulation tasks. HITL-D leverages a novel combination of diffusion-based policies and human control to provide autonomous end effector orientation updates conditioned on a scene point cloud and the Cartesian position of the end effector. This approach reduces the number of joystick control axes required, thereby lowering mental workload. In a multi-task user study with 12 participants, HITL-D reduced average task completion times by 40%, decreased perceived workload by 37%, and improved Likert-scale ratings for independence, intuitiveness, and confidence compared to traditional teleoperation methods. These results demonstrate that HITL-D effectively integrates human expertise with autonomous assistance, improving both objective and subjective aspects of teleoperation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_21460 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | HITL-D: Human In The Loop Diffusion Assisted Shared Control Zilka, Riley Khlynovskiy, Sergey Wang, Allie Jagersand, Martin Robotics Artificial Intelligence Human-Computer Interaction I.2 Autonomous manipulation systems have achieved remarkable capabilities, yet the integration of human expertise with diffusion-based policies in shared control remains relatively unexplored. In this paper, we propose Human-In-The-Loop Diffusion (HITL-D), a shared control framework that enhances user performance in multi-step, insertion, and fine manipulation tasks. HITL-D leverages a novel combination of diffusion-based policies and human control to provide autonomous end effector orientation updates conditioned on a scene point cloud and the Cartesian position of the end effector. This approach reduces the number of joystick control axes required, thereby lowering mental workload. In a multi-task user study with 12 participants, HITL-D reduced average task completion times by 40%, decreased perceived workload by 37%, and improved Likert-scale ratings for independence, intuitiveness, and confidence compared to traditional teleoperation methods. These results demonstrate that HITL-D effectively integrates human expertise with autonomous assistance, improving both objective and subjective aspects of teleoperation. |
| title | HITL-D: Human In The Loop Diffusion Assisted Shared Control |
| topic | Robotics Artificial Intelligence Human-Computer Interaction I.2 |
| url | https://arxiv.org/abs/2605.21460 |