X-Imitator: Spatial-Aware Imitation Learning via Bidirectional Action-Pose Interaction

X-Imitator: Spatial-Aware Imitation Learning via Bidirectional Action-Pose Interaction

Kai Xiong, Hongjie Fang, Lixin Yang, Cewu Lu · · 2026

Summary

Effectively handling the interplay between spatial perception and action generation remains a critical bottleneck in robotic manipulation. Existing methods typically treat spatial perception and action execution as decoupled or strictly...

Abstract Summary

Effectively handling the interplay between spatial perception and action generation remains a critical bottleneck in robotic manipulation. Existing methods typically treat spatial perception and action execution as decoupled or strictly unidirectional processes, fundamentally restricting a robot's ability to master complex manipulation tasks. To address this, we propose X-Imitator, a versatile dual-path framework that models spatial perception and action execution as a tightly coupled bidirectional loop. By reciprocally conditioning current pose predictions on past actions and vice versa, this framework enables continuous mutual refinement between spatial reasoning and action generation. This joint modeling exactly mimics human internal forward models. Designed as a modular architecture, the system can be seamlessly integrated into various visuomotor policies. Extensive experiments across 24 simulated and 3 real-world tasks demonstrate that our framework significantly outperforms both vanilla policies and prior methods utilizing explicit pose guidance. The code will be open sourced.

Key Points

  • Introduces X-Imitator, advancing imitation-learning research.
  • Detailed evaluation demonstrates effectiveness on benchmarks and real-world scenarios.
  • Provides insights into methodology and architectural choices.

Abstract

Effectively handling the interplay between spatial perception and action generation remains a critical bottleneck in robotic manipulation. Existing methods typically treat spatial perception and action execution as decoupled or strictly unidirectional processes, fundamentally restricting a robot’s abimitation-learningity to master complex manipulation tasks. To address this, we propose X-Imitator, a versatimitation-learninge dual-path framework that models spatial perception and action execution as a tightly coupled bidirectional loop. By reciprocally conditioning current pose predictions on past actions and vice versa, this framework enables continuous mutual refinement between spatial reasoning and action generation. This joint modeling exactly mimics human internal forward models. Designed as a modular architecture, the system can be seamlessly integrated into various visuomotor policies. Extensive experiments across 24 simulated and 3 real-world tasks demonstrate that our framework significantly outperforms both vanimitation-learningla policies and prior methods utimitation-learningizing explicit pose guidance. The code wimitation-learningl be open sourced.

Share

Related Papers

DROID: A Large-Scale In-the-Wild Robot Manipulation Dataset

DROID: A Large-Scale In-the-Wild Robot Manipulation Dataset

Alexander Khazatsky, Karl Pertsch, Suraj Nair et al. · RSS 2024 · Jul 2024

A 350-hour dataset of diverse real-world robot manipulation across 22 robots in 71 scenes, designed to train scalable and generalist imitation learning policies.

imitation-learning manipulation foundation-models
Code PDF Advanced
GitHub ★ 362 Code updated: May 2026
LeRobot: A Library for Real-World Robot Learning

LeRobot: A Library for Real-World Robot Learning

Remi Cadene, Simon Alibert, Alexander Soare et al. · NeurIPS 2024 Workshop · Jun 2024

Hugging Face's LeRobot is an open-source PyTorch framework providing pretrained models, datasets, and training scripts for imitation and reinforcement learning on real robots, lowering the entry barrier to robot learning.

imitation-learning rl foundation-models manipulation
Code PDF Beginner
GitHub ★ 24,333 Code updated: May 2026
Octo: An Open-Source Generalist Robot Policy

Octo: An Open-Source Generalist Robot Policy

Dibya Ghosh, Homer Walke, Karl Pertsch et al. · RSS 2024 · Jul 2024

Octo is a large open-source transformer-based generalist robot policy trained on 800k trajectories, supporting language-conditioned and goal-image-conditioned control across 9 robotic platforms with efficient fine-tuning on consumer GPUs.

foundation-models manipulation imitation-learning open-source
Code PDF Intermediate
GitHub ★ 1,652 Code updated: May 2026
Open X-Embodiment: Robotic Learning Datasets and RT-X Models

Open X-Embodiment: Robotic Learning Datasets and RT-X Models

Open X-Embodiment Collaboration · arXiv · Oct 2023

The largest collaborative robot learning dataset ever assembled, spanning 22 robots from 21 institutions, enabling training of generalist RT-X policies that exhibit positive cross-embodiment transfer.

imitation-learning foundation-models manipulation dataset
Code PDF Intermediate
GitHub ★ 1,853 Code updated: May 2026