mlsanityData to 14 Oct 2026

Prompt2Auto: From Motion Prompt to Automated Control via Geometry-Invariant One-Shot Gaussian Process Learning

arXiv 2509.14040 ↗First posted 17 Sep 2025cs.RO cs.AI cs.SYcited by 0

On the map: Offline RL and imitation learning from data → Robot learning from demonstration

Abstract (arXiv)

Learning from demonstration allows robots to acquire complex skills from human demonstrations, but conventional approaches often require large datasets and fail to generalize across coordinate transformations. In this paper, we propose Prompt2Auto, a geometry-invariant one-shot Gaussian process (GeoGP) learning framework that enables robots to perform human-guided automated control from a single motion prompt. A dataset-construction strategy based on coordinate transformations is introduced that enforces invariance to translation, rotation, and scaling, while supporting multi-step predictions. Moreover, GeoGP is robust to variations in the user's motion prompt and supports multi-skill autonomy. We validate the proposed approach through numerical simulations with the designed user graphical interface and two real-world robotic experiments, which demonstrate that the proposed method is effective, generalizes across tasks, and significantly reduces the demonstration burden. Project page is available at: https://prompt2auto.github.io

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Robot learning from demonstration, one of 13 directions in Offline RL and imitation learning from data. Builds on 9 of its 21 references that are in the map. First posted on arXiv 17 Sep 2025.

Builds on

9 of 21 references.

References · 9

  1. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,838
  2. 2015End-to-End Training of Deep Visuomotor Policiescited by 875
  3. 2022BC-Z: Zero-Shot Task Generalization with Robotic Imitation Learningcited by 523
  4. 2023A Survey of Imitation Learning: Algorithms, Recent Developments, and Challengescited by 148
  5. 2023HYDRA: Hybrid Robot Actions for Imitation Learningcited by 55
  6. 2017Streaming Sparse Gaussian Process Approximationscited by 41
  7. 2024ForceMimic: Force-Centric Imitation Learning with Force-Motion Capture System for Contact-Rich Manipulationcited by 36
  8. 2023Can Learning Deteriorate Control? Analyzing Computational Delays in Gaussian Process-Based Event-Triggered Online Learningcited by 8
  9. 2024Cooperative Learning with Gaussian Processes for Euler-Lagrange Systems Tracking Control under Switching Topologiescited by 4

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2020Gaussian-Process-based Robot Learning from Demonstration
  2. 2019Learning from demonstration with model-based Gaussian process
  3. 2025Generalizing Robot Trajectories from Single-Context Human Demonstrations: A Probabilistic Approach
  4. 2025MOVE: A Simple Motion-Based Data Collection Paradigm for Spatial Generalization in Robotic Manipulation
  5. 2020Task-Adaptive Robot Learning from Demonstration with Gaussian Process Models under Replication
  6. 2024Learning Generalizable 3D Manipulation With 10 Demonstrations
  7. 2018Auto-conditioned Recurrent Mixture Density Networks for Learning Generalizable Robot Skills
  8. 2025The Unreasonable Effectiveness of Discrete-Time Gaussian Process Mixtures for Robot Policy Learning