# KiRAS: Keyframe Guided Self-Imitation for Robust and Adaptive Skill Learning in Quadruped Robots

arXiv: https://arxiv.org/abs/2603.15179 · HTML: https://mlsanity.com/paper/2603.15179/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Robot manipulation imitation learning, one of 13 directions in Offline RL and imitation learning from data.
- Builds on 24 of its 27 references that are in the map.
- First posted on arXiv 16 Mar 2026.

## Placed in

- Offline RL and imitation learning from data / [Robot manipulation imitation learning](https://mlsanity.com/offlinerl/robot-manipulation-imitation-learning/)

## Builds on (most cited first, 24 of 24)

- [Generative Adversarial Imitation Learning](https://mlsanity.com/paper/1606.03476/) (2016)
- [Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning](https://mlsanity.com/paper/2108.10470/) (2021)
- [Learning to Learn without Forgetting by Maximizing Transfer and Minimizing Interference](https://mlsanity.com/paper/1810.11910/) (2018)
- [Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learning](https://mlsanity.com/paper/2109.11978/) (2021)
- [RMA: Rapid Motor Adaptation for Legged Robots](https://mlsanity.com/paper/2107.04034/) (2021)
- [Learning robust perceptive locomotion for quadrupedal robots in the wild](https://mlsanity.com/paper/2201.08117/) (2022)
- [Self-Imitation Learning](https://mlsanity.com/paper/1806.05635/) (2018)
- [Walk These Ways: Tuning Robot Control for Generalization with Multiplicity of Behavior](https://mlsanity.com/paper/2212.03238/) (2022)
- [Concurrent Training of a Control Policy and a State Estimator for Dynamic and Robust Legged Locomotion](https://mlsanity.com/paper/2202.05481/) (2022)
- [DreamWaQ: Learning Robust Quadrupedal Locomotion With Implicit Terrain Imagination via Deep Reinforcement Learning](https://mlsanity.com/paper/2301.10602/) (2023)
- [Learning Agile Skills via Adversarial Imitation of Rough Partial Demonstrations](https://mlsanity.com/paper/2206.11693/) (2022)
- [Hybrid Internal Model: Learning Agile Legged Locomotion with Simulated Robot Response](https://mlsanity.com/paper/2312.11460/) (2023)
- [Advanced Skills through Multiple Adversarial Motion Priors in Reinforcement Learning](https://mlsanity.com/paper/2203.14912/) (2022)
- [Lifelike Agility and Play in Quadrupedal Robots using Reinforcement Learning and Generative Pre-trained Models](https://mlsanity.com/paper/2308.15143/) (2023)
- [OPT-Mimic: Imitation of Optimized Trajectories for Dynamic Quadruped Behaviors](https://mlsanity.com/paper/2210.01247/) (2022)
- [CaT: Constraints as Terminations for Legged Locomotion Reinforcement Learning](https://mlsanity.com/paper/2403.18765/) (2024)
- [MoE-Loco: Mixture of Experts for Multitask Locomotion](https://mlsanity.com/paper/2503.08564/) (2025)
- [An Open Torque-Controlled Modular Robot Architecture for Legged Locomotion Research](https://mlsanity.com/paper/1910.00093/) (2019)
- [Versatile Skill Control via Self-supervised Adversarial Imitation of Unlabeled Mixed Motions](https://mlsanity.com/paper/2209.07899/) (2022)
- [FLD: Fourier Latent Dynamics for Structured Motion Representation and Learning](https://mlsanity.com/paper/2402.13820/) (2024)
- [RobotKeyframing: Learning Locomotion with High-Level Objectives via Mixture of Dense and Sparse Rewards](https://mlsanity.com/paper/2407.11562/) (2024)
- [Human2LocoMan: Learning Versatile Quadrupedal Manipulation with Human Pretraining](https://mlsanity.com/paper/2506.16475/) (2025)
- [Motion Priors Reimagined: Adapting Flat-Terrain Skills for Complex Quadruped Mobility](https://mlsanity.com/paper/2505.16084/) (2025)
- [Continuous Control of Diverse Skills in Quadruped Robots Without Complete Expert Datasets](https://mlsanity.com/paper/2503.03476/) (2025)

## Abstract (arXiv)

With advances in reinforcement learning and imitation learning, quadruped robots can acquire diverse skills within a single policy by imitating multiple skill-specific datasets. However, the lack of datasets on complex terrains limits the ability of such multi-skill policies to generalize effectively in unstructured environments. Inspired by animation, we adopt keyframes as minimal and universal skill representations, relaxing dataset constraints and enabling the integration of terrain adaptability with skill diversity. We propose Keyframe Guided Self-Imitation for Robust and Adaptive Skill Learning (KiRAS), an end-to-end framework for acquiring and transitioning between diverse skill primitives on complex terrains. KiRAS first learns diverse skills on flat terrain through keyframe-guided self-imitation, eliminating the need for expert datasets; then continues training the same policy network on rough terrains to enhance robustness. To eliminate catastrophic forgetting, a proficiency-based Skill Initialization Technique is introduced. Experiments on Solo-8 and Unitree Go1 robots show that KiRAS enables robust skill acquisition and smooth transitions across challenging terrains. This framework demonstrates its potential as a lightweight platform for multi-skill generation and dataset collection. It further enables flexible skill transitions that enhance locomotion on challenging terrains.