# TERP: Reliable Planning in Uneven Outdoor Environments using Deep Reinforcement Learning

arXiv: https://arxiv.org/abs/2109.05120 · HTML: https://mlsanity.com/paper/2109.05120/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Sep 2021.

## Built on by (most cited first, 20 of 20)

- [ScaTE: A Scalable Framework for Self-Supervised Traversability Estimation in Unstructured Environments](https://mlsanity.com/paper/2209.06522/) (2022)
- [Safe Reinforcement Learning of Dynamic High-Dimensional Robotic Tasks: Navigation, Manipulation, Interaction](https://mlsanity.com/paper/2209.13308/) (2022)
- [Safe Reinforcement Learning on the Constraint Manifold: Theory and Applications](https://mlsanity.com/paper/2404.09080/) (2024)
- [Robot Navigation Using Physically Grounded Vision-Language Models in Outdoor Environments](https://mlsanity.com/paper/2409.20445/) (2024)
- [CREStE: Scalable Mapless Navigation with Internet Scale Priors and Counterfactual Guidance](https://mlsanity.com/paper/2503.03921/) (2025)
- [VINet: Visual and Inertial-based Terrain Classification and Adaptive Navigation over Unknown Terrain](https://mlsanity.com/paper/2209.07725/) (2022)
- [CoverNav: Cover Following Navigation Planning in Unstructured Outdoor Environment with Deep Reinforcement Learning](https://mlsanity.com/paper/2308.06594/) (2023)
- [Autonomous Driving in Unstructured Environments: How Far Have We Come?](https://mlsanity.com/paper/2410.07701/) (2024)
- [Adaptive Diffusion Terrain Generator for Autonomous Uneven Terrain Navigation](https://mlsanity.com/paper/2410.10766/) (2024)
- [SOCIALGYM 2.0: Simulator for Multi-Agent Social Robot Navigation in Shared Human Spaces](https://mlsanity.com/paper/2303.05584/) (2023)
- [Safe Navigation in Unstructured Environments by Minimizing Uncertainty in Control and Perception](https://mlsanity.com/paper/2306.14601/) (2023)
- [RE-MOVE: An Adaptive Policy Design for Robotic Navigation Tasks in Dynamic Environments via Language-Based Feedback](https://mlsanity.com/paper/2303.07622/) (2023)
- [Vision based Crop Row Navigation under Varying Field Conditions in Arable Fields](https://mlsanity.com/paper/2209.14003/) (2022)
- [VAPOR: Legged Robot Navigation in Outdoor Vegetation Using Offline Reinforcement Learning](https://mlsanity.com/paper/2309.07832/) (2023)
- [CAMEL: Learning Cost-maps Made Easy for Off-road Driving](https://mlsanity.com/paper/2209.12413/) (2022)
- [Confidence-Controlled Exploration: Efficient Sparse-Reward Policy Learning for Robot Navigation](https://mlsanity.com/paper/2306.06192/) (2023)
- [METAVerse: Meta-Learning Traversability Cost Map for Off-Road Navigation](https://mlsanity.com/paper/2307.13991/) (2023)
- [NavG: Risk-Aware Navigation in Crowded Environments Based on Reinforcement Learning with Guidance Points](https://mlsanity.com/paper/2503.02111/) (2025)
- [ADEPT: Adaptive Diffusion Environment for Policy Transfer Sim-to-Real](https://mlsanity.com/paper/2506.01759/) (2025)
- [Learning Smooth State-Dependent Traversability from Dense Point Clouds](https://mlsanity.com/paper/2506.04362/) (2025)

## Abstract (arXiv)

We present a novel method for reliable robot navigation in uneven outdoor terrains. Our approach employs a novel fully-trained Deep Reinforcement Learning (DRL) network that uses elevation maps of the environment, robot pose, and goal as inputs to compute an attention mask of the environment. The attention mask is used to identify reduced stability regions in the elevation map and is computed using channel and spatial attention modules and a novel reward function. We continuously compute and update a navigation cost-map that encodes the elevation information or the level-of-flatness of the terrain using the attention mask. We then generate locally least-cost waypoints on the cost-map and compute the final dynamically feasible trajectory using another DRL-based method. Our approach guarantees safe, locally least-cost paths and dynamically feasible robot velocities in uneven terrains. We observe an increase of 35.18% in terms of success rate and, a decrease of 26.14% in the cumulative elevation gradient of the robot's trajectory compared to prior navigation methods in high-elevation regions. We evaluate our method on a Husky robot in real-world uneven terrains (~ 4m of elevation gain) and demonstrate its benefits.