# Robot Crash Course: Learning Soft and Stylized Falling

arXiv: https://arxiv.org/abs/2511.10635 · HTML: https://mlsanity.com/paper/2511.10635/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 37 references that are in the map.
- First posted on arXiv 13 Nov 2025.

## Builds on (most cited first, 22 of 22)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)](https://mlsanity.com/paper/1511.07289/) (2015)
- [Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learning](https://mlsanity.com/paper/2109.11978/) (2021)
- [Asymmetric Actor Critic for Image-Based Robot Learning](https://mlsanity.com/paper/1710.06542/) (2017)
- [Real-World Humanoid Locomotion with Reinforcement Learning](https://mlsanity.com/paper/2303.03381/) (2023)
- [BeyondMimic: From Motion Tracking to Versatile Humanoid Control via Guided Diffusion](https://mlsanity.com/paper/2508.08241/) (2025)
- [Humanoid Parkour Learning](https://mlsanity.com/paper/2406.10759/) (2024)
- [Residual Force Control for Agile Human Behavior Imitation and Extended Motion Synthesis](https://mlsanity.com/paper/2006.07364/) (2020)
- [Learning Vision-Based Bipedal Locomotion for Challenging Terrain](https://mlsanity.com/paper/2309.14594/) (2023)
- [Optimization-Based Control for Dynamic Legged Robots](https://mlsanity.com/paper/2211.11644/) (2022)
- [Learning Getting-Up Policies for Real-World Humanoid Robots](https://mlsanity.com/paper/2502.12152/) (2025)
- [Learning Arm-Assisted Fall Damage Reduction and Recovery for Legged Mobile Manipulators](https://mlsanity.com/paper/2303.05486/) (2023)
- [RobotKeyframing: Learning Locomotion with High-Level Objectives via Mixture of Dense and Sparse Rewards](https://mlsanity.com/paper/2407.11562/) (2024)
- [Design and Control of a Bipedal Robotic Character](https://mlsanity.com/paper/2501.05204/) (2025)
- [Learning to Get Up](https://mlsanity.com/paper/2205.00307/) (2022)
- [AMOR: Adaptive Character Control through Multi-Objective Reinforcement Learning](https://mlsanity.com/paper/2505.23708/) (2025)
- [Online Non-linear Centroidal MPC for Humanoid Robot Locomotion with Step Adjustment](https://mlsanity.com/paper/2203.04489/) (2022)
- [Reward Scale Robustness for Proximal Policy Optimization via DreamerV3 Tricks](https://mlsanity.com/paper/2310.17805/) (2023)
- [Guardians as You Fall: Active Mode Transition for Safe Falling](https://mlsanity.com/paper/2310.04828/) (2023)
- [Learning a Unified Control Policy for Safe Falling](https://mlsanity.com/paper/1703.02905/) (2017)
- [Autonomous Human-Robot Interaction via Operator Imitation](https://mlsanity.com/paper/2504.02724/) (2025)
- [Learning to Walk in Costume: Adversarial Motion Priors for Aesthetically Constrained Humanoids](https://mlsanity.com/paper/2509.05581/) (2025)

## Built on by (most cited first, 4 of 4)

- [CoCo-InEKF: State Estimation with Learned Contact Covariances in Dynamic, Contact-Rich Scenarios](https://mlsanity.com/paper/2605.15122/) (2026)
- [Constrained Whole-Body Tracking for Humanoid Robots](https://mlsanity.com/paper/2606.00374/) (2026)
- [Bicycle Acrobatics with Reinforcement Learning](https://mlsanity.com/paper/2608.00880/) (2026)
- [Humanoid Safe Stop via Learned Stoppability Value](https://mlsanity.com/paper/2609.02358/) (2026)

## Abstract (arXiv)

Despite recent advances in robust locomotion, bipedal robots operating in the real world remain at risk of falling. While most research focuses on preventing such events, we instead concentrate on the phenomenon of falling itself. Specifically, we aim to reduce physical damage to the robot while providing users with control over a robot's end pose. To this end, we propose a robot agnostic reward function that balances the achievement of a desired end pose with impact minimization and the protection of critical robot parts during reinforcement learning. To make the policy robust to a broad range of initial falling conditions and to enable the specification of an arbitrary and unseen end pose at inference time, we introduce a simulation-based sampling strategy of initial and end poses. Through simulated and real-world experiments, our work demonstrates that even bipedal robots can perform controlled, soft falls.