# Safety-Aware Imitation Learning via MPC-Guided Disturbance Injection

arXiv: https://arxiv.org/abs/2508.03129 · HTML: https://mlsanity.com/paper/2508.03129/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Imitation learning methods, one of 13 directions in Offline RL and imitation learning from data.
- Builds on 23 of its 34 references that are in the map.
- First posted on arXiv 5 Aug 2025.

## Placed in

- Offline RL and imitation learning from data / [Imitation learning methods](https://mlsanity.com/offlinerl/imitation-learning-methods/)

## Builds on (most cited first, 23 of 23)

- [Generative Adversarial Imitation Learning](https://mlsanity.com/paper/1606.03476/) (2016)
- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning](https://mlsanity.com/paper/2108.10470/) (2021)
- [Learning Robust Rewards with Adversarial Inverse Reinforcement Learning](https://mlsanity.com/paper/1710.11248/) (2017)
- [End-to-end Driving via Conditional Imitation Learning](https://mlsanity.com/paper/1710.02410/) (2017)
- [Control Barrier Functions: Theory and Applications](https://mlsanity.com/paper/1903.11199/) (2019)
- [Learning Agile Robotic Locomotion Skills by Imitating Animals](https://mlsanity.com/paper/2004.00784/) (2020)
- [One-Shot Visual Imitation Learning via Meta-Learning](https://mlsanity.com/paper/1709.04905/) (2017)
- [Hamilton-Jacobi Reachability: A Brief Overview and Recent Advances](https://mlsanity.com/paper/1709.07523/) (2017)
- [DART: Noise Injection for Robust Imitation Learning](https://mlsanity.com/paper/1703.09327/) (2017)
- [Walk These Ways: Tuning Robot Control for Generalization with Multiplicity of Behavior](https://mlsanity.com/paper/2212.03238/) (2022)
- [The Safety Filter: A Unified View of Safety-Critical Control in Autonomous Systems](https://mlsanity.com/paper/2309.05837/) (2023)
- [NeRF in the Palm of Your Hand: Corrective Augmentation for Robotics via Novel-View Synthesis](https://mlsanity.com/paper/2301.08556/) (2023)
- [Robust Control Barrier-Value Functions for Safety-Critical Control](https://mlsanity.com/paper/2104.02808/) (2021)
- [CCIL: Continuity-based Data Augmentation for Corrective Imitation Learning](https://mlsanity.com/paper/2310.12972/) (2023)
- [IntervenGen: Interventional Data Generation for Robust and Data-Efficient Robot Imitation Learning](https://mlsanity.com/paper/2405.01472/) (2024)
- [TaSIL: Taylor Series Imitation Learning](https://mlsanity.com/paper/2205.14812/) (2022)
- [An Efficient Reachability-Based Framework for Provably Safe Autonomous Navigation in Unknown Environments](https://mlsanity.com/paper/1905.00532/) (2019)
- [Gameplay Filters: Robust Zero-Shot Safety through Adversarial Imagination](https://mlsanity.com/paper/2405.00846/) (2024)
- [Stable-BC: Controlling Covariate Shift with Stable Behavior Cloning](https://mlsanity.com/paper/2408.06246/) (2024)
- [L1-Adaptive MPPI Architecture for Robust and Agile Control of Multirotors](https://mlsanity.com/paper/2004.00152/) (2020)
- [SAFE-GIL: SAFEty Guided Imitation Learning for Robotic Systems](https://mlsanity.com/paper/2404.05249/) (2024)
- [DualGuard MPPI: Safe and Performant Optimal Control by Combining Sampling-Based MPC and Hamilton-Jacobi Reachability](https://mlsanity.com/paper/2502.01924/) (2025)

## Built on by (most cited first, 1 of 1)

- [Where Small Mistakes Matter: Quantifying State Criticality for Imitation Learning](https://mlsanity.com/paper/2610.17709/) (2026)

## Abstract (arXiv)

Imitation Learning has provided a promising approach to learning complex robot behaviors from expert demonstrations. However, learned policies can make errors that lead to safety violations, which limits their deployment in safety-critical applications. We propose MPC-SafeGIL, a design-time approach that enhances the safety of imitation learning by injecting adversarial disturbances during expert demonstrations. This exposes the expert to a broader range of safety-critical scenarios and allows the imitation policy to learn robust recovery behaviors. Our method uses sampling-based Model Predictive Control (MPC) to approximate worst-case disturbances, making it scalable to high-dimensional and black-box dynamical systems. In contrast to prior work that relies on analytical models or interactive experts, MPC-SafeGIL integrates safety considerations directly into data collection. We validate our approach through extensive simulations including quadruped locomotion and visuomotor navigation and real-world experiments on a quadrotor, demonstrating improvements in both safety and task performance. See our website here: https://leqiu2003.github.io/MPCSafeGIL/