# Improving Generalization in Reinforcement Learning Training Regimes for Social Robot Navigation

arXiv: https://arxiv.org/abs/2308.14947 · HTML: https://mlsanity.com/paper/2308.14947/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 37 references that are in the map.
- First posted on arXiv 29 Aug 2023.

## Builds on (most cited first, 10 of 10)

- [Social Attention: Modeling Attention in Human Crowds](https://mlsanity.com/paper/1710.04689/) (2017)
- [Crowd-Robot Interaction: Crowd-aware Robot Navigation with Attention-based Deep Reinforcement Learning](https://mlsanity.com/paper/1809.08835/) (2018)
- [Socially Aware Motion Planning with Deep Reinforcement Learning](https://mlsanity.com/paper/1703.08862/) (2017)
- [Decentralized Non-communicating Multiagent Collision Avoidance with Deep Reinforcement Learning](https://mlsanity.com/paper/1609.07845/) (2016)
- [Motion Planning Among Dynamic, Decision-Making Agents with Deep Reinforcement Learning](https://mlsanity.com/paper/1805.01956/) (2018)
- [Core Challenges of Social Robot Navigation: A Survey](https://mlsanity.com/paper/2103.05668/) (2021)
- [Socially Compliant Navigation through Raw Depth Inputs with Generative Adversarial Imitation Learning](https://mlsanity.com/paper/1710.02543/) (2017)
- [Risk-Sensitive Sequential Action Control with Multi-Modal Human Trajectory Forecasting for Safe Crowd-Robot Interaction](https://mlsanity.com/paper/2009.05702/) (2020)
- [Deep-Learned Collision Avoidance Policy for Distributed Multi-Agent Navigation](https://mlsanity.com/paper/1609.06838/) (2016)
- [Deep Social Force](https://mlsanity.com/paper/2109.12081/) (2021)

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

- [Don't Freeze, Don't Crash: Extending the Safe Operating Range of Neural Navigation in Dense Crowds](https://mlsanity.com/paper/2603.06729/) (2026)

## Abstract (arXiv)

In order for autonomous mobile robots to navigate in human spaces, they must abide by our social norms. Reinforcement learning (RL) has emerged as an effective method to train sequential decision-making policies that are able to respect these norms. However, a large portion of existing work in the field conducts both RL training and testing in simplistic environments. This limits the generalization potential of these models to unseen environments, and the meaningfulness of their reported results. We propose a method to improve the generalization performance of RL social navigation methods using curriculum learning. By employing multiple environment types and by modeling pedestrians using multiple dynamics models, we are able to progressively diversify and escalate difficulty in training. Our results show that the use of curriculum learning in training can be used to achieve better generalization performance than previous training methods. We also show that results presented in many existing state-of-the-art RL social navigation works do not evaluate their methods outside of their training environments, and thus do not reflect their policies' failure to adequately generalize to out-of-distribution scenarios. In response, we validate our training approach on larger and more crowded testing environments than those used in training, allowing for more meaningful measurements of model performance.