ManiLong-Shot: Interaction-Aware One-Shot Imitation Learning for Long-Horizon Manipulation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914207144869888 |
|---|---|
| author | Chen, Zixuan Gao, Chongkai Shao, Lin Shi, Jieqi Huo, Jing Gao, Yang |
| author_facet | Chen, Zixuan Gao, Chongkai Shao, Lin Shi, Jieqi Huo, Jing Gao, Yang |
| contents | One-shot imitation learning (OSIL) offers a promising way to teach robots new skills without large-scale data collection. However, current OSIL methods are primarily limited to short-horizon tasks, thus limiting their applicability to complex, long-horizon manipulations. To address this limitation, we propose ManiLong-Shot, a novel framework that enables effective OSIL for long-horizon prehensile manipulation tasks. ManiLong-Shot structures long-horizon tasks around physical interaction events, reframing the problem as sequencing interaction-aware primitives instead of directly imitating continuous trajectories. This primitive decomposition can be driven by high-level reasoning from a vision-language model (VLM) or by rule-based heuristics derived from robot state changes. For each primitive, ManiLong-Shot predicts invariant regions critical to the interaction, establishes correspondences between the demonstration and the current observation, and computes the target end-effector pose, enabling effective task execution. Extensive simulation experiments show that ManiLong-Shot, trained on only 10 short-horizon tasks, generalizes to 20 unseen long-horizon tasks across three difficulty levels via one-shot imitation, achieving a 22.8% relative improvement over the SOTA. Additionally, real-robot experiments validate ManiLong-Shot's ability to robustly execute three long-horizon manipulation tasks via OSIL, confirming its practical applicability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_16302 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ManiLong-Shot: Interaction-Aware One-Shot Imitation Learning for Long-Horizon Manipulation Chen, Zixuan Gao, Chongkai Shao, Lin Shi, Jieqi Huo, Jing Gao, Yang Robotics One-shot imitation learning (OSIL) offers a promising way to teach robots new skills without large-scale data collection. However, current OSIL methods are primarily limited to short-horizon tasks, thus limiting their applicability to complex, long-horizon manipulations. To address this limitation, we propose ManiLong-Shot, a novel framework that enables effective OSIL for long-horizon prehensile manipulation tasks. ManiLong-Shot structures long-horizon tasks around physical interaction events, reframing the problem as sequencing interaction-aware primitives instead of directly imitating continuous trajectories. This primitive decomposition can be driven by high-level reasoning from a vision-language model (VLM) or by rule-based heuristics derived from robot state changes. For each primitive, ManiLong-Shot predicts invariant regions critical to the interaction, establishes correspondences between the demonstration and the current observation, and computes the target end-effector pose, enabling effective task execution. Extensive simulation experiments show that ManiLong-Shot, trained on only 10 short-horizon tasks, generalizes to 20 unseen long-horizon tasks across three difficulty levels via one-shot imitation, achieving a 22.8% relative improvement over the SOTA. Additionally, real-robot experiments validate ManiLong-Shot's ability to robustly execute three long-horizon manipulation tasks via OSIL, confirming its practical applicability. |
| title | ManiLong-Shot: Interaction-Aware One-Shot Imitation Learning for Long-Horizon Manipulation |
| topic | Robotics |
| url | https://arxiv.org/abs/2512.16302 |