# PACER: Preference-conditioned All-terrain Costmap Generation

arXiv: https://arxiv.org/abs/2410.23488 · HTML: https://mlsanity.com/paper/2410.23488/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 37 references that are in the map.
- First posted on arXiv 30 Oct 2024.

## Builds on (most cited first, 19 of 19)

- [Socially Compliant Navigation Dataset (SCAND): A Large-Scale Dataset of Demonstrations for Social Navigation](https://mlsanity.com/paper/2203.15041/) (2022)
- [RELLIS-3D Dataset: Data, Benchmarks and Analysis](https://mlsanity.com/paper/2011.12954/) (2020)
- [BADGR: An Autonomous Self-Supervised Learning-Based Navigation System](https://mlsanity.com/paper/2002.05700/) (2020)
- [Motion Planning and Control for Mobile Robot Navigation Using Machine Learning: a Survey](https://mlsanity.com/paper/2011.13112/) (2020)
- [GANav: Efficient Terrain Segmentation for Robot Navigation in Unstructured Outdoor Environments](https://mlsanity.com/paper/2103.04233/) (2021)
- [Elevation Mapping for Locomotion and Navigation using GPU](https://mlsanity.com/paper/2204.12876/) (2022)
- [Self-Supervised Visual Terrain Classification from Unsupervised Acoustic Feature Learning](https://mlsanity.com/paper/1912.03227/) (2019)
- [TerrainNet: Visual Modeling of Complex Terrain for High-speed, Off-road Navigation](https://mlsanity.com/paper/2303.15771/) (2023)
- [LaND: Learning to Navigate from Disengagements](https://mlsanity.com/paper/2010.04689/) (2020)
- [Visual Representation Learning for Preference-Aware Path Planning](https://mlsanity.com/paper/2109.08968/) (2021)
- [Rethinking Social Robot Navigation: Leveraging the Best of Two Worlds](https://mlsanity.com/paper/2309.13466/) (2023)
- [From the Desks of ROS Maintainers: A Survey of Modern & Capable Mobile Robotics Algorithms in the Robot Operating System 2](https://mlsanity.com/paper/2307.15236/) (2023)
- [STERLING: Self-Supervised Terrain Representation Learning from Unconstrained Robot Experience](https://mlsanity.com/paper/2309.15302/) (2023)
- [RCA: Ride Comfort-Aware Visual Navigation via Self-Supervised Learning](https://mlsanity.com/paper/2207.14460/) (2022)
- [Wait, That Feels Familiar: Learning to Extrapolate Human Preferences for Preference Aligned Path Planning](https://mlsanity.com/paper/2309.09912/) (2023)
- [Incorporating Human Domain Knowledge into Large Scale Cost Function Learning](https://mlsanity.com/paper/1612.04318/) (2016)
- [Robot navigation from human demonstration: learning control behaviors with environment feature maps](https://mlsanity.com/paper/2205.03364/) (2022)
- [Autonomous Ground Navigation in Highly Constrained Spaces: Lessons learned from The BARN Challenge at ICRA 2022](https://mlsanity.com/paper/2208.10473/) (2022)
- [Autonomous Ground Navigation in Highly Constrained Spaces: Lessons learned from The 2nd BARN Challenge at ICRA 2023](https://mlsanity.com/paper/2308.03205/) (2023)

## Built on by (most cited first, 5 of 5)

- [CREStE: Scalable Mapless Navigation with Internet Scale Priors and Counterfactual Guidance](https://mlsanity.com/paper/2503.03921/) (2025)
- [VENTURA: Adapting Image Diffusion Models for Unified Task Conditioned Navigation](https://mlsanity.com/paper/2510.01388/) (2025)
- [OVerSeeC: Open-Vocabulary Costmap Generation from Satellite Images and Natural Language](https://mlsanity.com/paper/2602.18606/) (2026)
- [MoMo: Conditioned Contrastive Representation Learning for Preference-Modulated Planning](https://mlsanity.com/paper/2605.08512/) (2026)
- [From Language Priors to Field Adaptation: Preference Learning for Traversability Estimation](https://mlsanity.com/paper/2610.02974/) (2026)

## Abstract (arXiv)

In autonomous robot navigation, terrain cost assignment is typically performed using a semantics-based paradigm in which terrain is first labeled using a pre-trained semantic classifier and costs are then assigned according to a user-defined mapping between label and cost. While this approach is rapidly adaptable to changing user preferences, only preferences over the types of terrain that are already known by the semantic classifier can be expressed. In this paper, we hypothesize that a machine-learning-based alternative to the semantics-based paradigm above will allow for rapid cost assignment adaptation to preferences expressed over new terrains at deployment time without the need for additional training. To investigate this hypothesis, we introduce and study PACER, a novel approach to costmap generation that accepts as input a single birds-eye view (BEV) image of the surrounding area along with a user-specified preference context and generates a corresponding BEV costmap that aligns with the preference context. Using both real and synthetic data along with a combination of proposed training tasks, we find that PACER is able to adapt quickly to new user preferences while also exhibiting better generalization to novel terrains compared to both semantics-based and representation-learning approaches.