General Information
The Resilient Off-road Autonomous Robotics (ROAR) builds on the success of the ICRA 2024 Workshop on Resilient Off-road Autonomy with a refined and sharpened focus on tackling the unique challenges posed by unstructured, unpredictable, and dynamic off-road environments. These settings require advanced capabilities in perception, localization, planning, and control - areas where conventional autonomous systems often fall short.
Beyond canonical topics such as robust navigation in GPS-denied areas and terrain-adaptive control strategies, the workshop will explore cutting-edge advancements. These include multi-modal perception using novel sensor modalities (e.g., radar for robust navigation, thermal cameras for visibility in low-light conditions, and event cameras for high-speed motion detection). Advanced scene representation techniques, such as NeRF for photorealistic scene reconstruction and 3D Gaussian Splatting for efficient map representations, will also be discussed. The workshop will highlight online learning approaches to enable real-time adaptation to dynamic off-road environments and examine how foundation models can contribute to generalizable and scalable off-road autonomy. Additionally, a significant focus will be placed on practical strategies for creating comprehensive datasets and establishing robust benchmarks tailored to off-road environments, emphasizing the importance of large-scale, high-quality data for method development and training.
This will be a full-day workshop. It will feature a mix of presentations, open panel discussions, and an invited poster session. The workshop targets academics and researchers in robotics, AI, and machine learning, focusing on autonomous systems for off-road applications. It also engages industry professionals from agriculture, mining, defense, and planetary exploration, along with representatives from organizations like NASA and DARPA.
π’ Updates!
-
π Online Access: Join the workshop via Zoom Link/Zoom Invitation
-
ποΈ Poster Schedule Released: Please check the schedule here
-
π Accepted Papers: All accepted papers are now available on OpenReview
π Location
-
Workshop Location: Olin Hall of Engineering (OHE) Room 100B
-
Poster Location: Epstein Family Plaza
Awards
Thank you for the sponsor from Field AI. The following prizes will be presented:
- Best Paper Award β USD 500
- Best Paper Runner-Up β USD 250
- Best Poster Award β USD 500
- Best Poster Runner-Up β USD 250
Speakers and Panelists
Schedule
| Time ββββββ | Session |
|---|---|
| 08:15-08:30 | Opening Remarks |
| 08:30-09:00 | Keynote Talk 1: Radar Localization for Driving Off the Road (Tim Barfoot) |
| 09:00-09:30 | Keynote Talk 2: Learning Robust Computer Vision in Unknown Environments (Marija PopoviΔ) |
| 09:30-09:45 | Oral Presentation 1: Meta-Learning Online Dynamics Model Adaptation in Off-Road Autonomous Driving |
| 09:45-10:30 | Coffee break and Poster Session |
| 10:30-11:00 | Keynote Talk 3: Robot Navigation in Complex Indoor and Outdoor Environment (Dinesh Manocha) |
| 11:00-11:30 | Keynote Talk 4: Future Surface and Aerial Mobility for Mars Exploration (Masahiro (Hiro) Ono) |
| 11:30-11:45 | Oral Presentation 2: OVerSeeC β Open-Vocabulary CostMap Generation from Satellite Images and Natural Language |
| 11:45-12:00 | Oral Presentation 3: Online Adaptation of Terrain-Aware Dynamics for Planning in Unstructured Environments |
| 12:00-12:15 | Oral Presentation 4: Kinodynamic Motion Planning for Mobile Robot Navigation across Inconsistent World Models |
| 12:15-12:30 | Oral Presentation 5: Long Range Navigator (LRN): Extending robot planning horizons beyond metric maps |
| 12:30-14:00 | Lunch Break. |
| 14:00-14:30 | Keynote Talk 5: Off Road Autonomy in Feature Sparse Environments (Anton Netchaev) |
| 14:30-15:00 | Keynote Talk 6: Thoughts on Machine Learning and Off-Road Ground Vehicle Autonomy (Byron Boots) |
| 15:00-15:30 | Keynote Talk 7: Reading the Dirt: Learning What Matters (and What Doesnβt) for Off-Road Navigation (Joydeep Biswas) |
| 15:30-16:00 | Coffee break and Poster Session |
| 16:00-16:30 | Keynote Talk 8: Spatial AI and ML for Autonomous Robot Navigation in Rough Terrains (Hyun Myung) |
| 16:30-17:00 | Keynote Talk 9: Robotic Autonomy in Complex Environments with Resiliency (Stuart Young) |
| 17:00-17:50 | Panel Discussion: Foundation Models and Large Language Models in Off-Road Robotics: Hype or Reality? |
| 17:50-18:00 | Closing Remarks. |
Poster Sessions
Session A - 9:30-10:20 AM (Epstein Plaza)
Long Range Navigator (LRN): Extending Robot Planning Horizons Beyond Metric Maps Matt Schmittle, Rohan Baijal, Nathan Hatch, Rosario Scalise, Mateo Guaman Castro, Sidharth Talia, Khimya Khetarpal, Byron Boots, Siddhartha Srinivasa
Kinodynamic Motion Planning for Mobile Robot Navigation Across Inconsistent World Models Eric R. Damm, Thomas Howard
Online Adaptation of Terrain-Aware Dynamics for Planning in Unstructured Environments William Ward, Sarah Etter, Tyler Ingebrand, Christian Ellis, Adam Thorpe, Ufuk Topcu
OVerSeeC β Open-Vocabulary CostMap Generation from Satellite Images and Natural Language Rwik Rana, Jesse Quattrociocchi, Dongmyeong Lee, Christian Ellis, Amanda Adkins, Adam Uccello, Garrett Warnell, Joydeep Biswas
Meta-Learning Online Dynamics Model Adaptation in Off-Road Autonomous Driving Jacob Levy, Jason Gibson, Bogdan Vlahov, Erica Tevere, Evangelos Theodorou, David Fridovich-Keil, Patrick Spieler
GND: An Outdoor Global Navigation Dataset with Multi-Modal Perception and Traversability Jing Liang, Dibyendu Das, Daeun Song, MD Nahid Hasan Shuvo, Sina Mansouri, Dinesh Manocha, Xuesu Xiao
Verti-Bench: A General and Scalable Off-Road Mobility Benchmark for Vertically Challenging Terrain Tong Xu, Chenhui Pan, Madhan B. Rao, Aniket Datar, Anuj Pokhrel, Yuanjie Lu, Xuesu Xiao
VertiSelector: Automatic Curriculum Learning for Wheeled Mobility on Vertically Challenging Terrain Tong Xu, Chenhui Pan, Xuesu Xiao
Learning Smooth State-Dependent Traversability from Dense Point Clouds Zihao Dong, Alan Papalia, Leonard Jung, Alenna Spiro, Philip R Osteen, Christa S. Robison, Michael Everett
Session C 3:00-3:50 PM (Epstein Plaza)
Traverse the Non-Traversable: Estimating Traversability for Wheeled Mobility on Vertically Challenging Terrain Aniket Datar, Chenhui Pan, Anuj Pokhrel, Matthew Choulas, Mohammad Nazeri, Tong Xu, Xuesu Xiao
M2P2: A Multi-Modal Passive Perception Dataset for Off-Road Mobility in Extreme Low-Light Conditions Aniket Datar, Anuj Pokhrel, Mohammad Nazeri, Madhan B. Rao, Harsh Rangwala, Chenhui Pan, Yufan Zhang, Andre Harrison, Maggie Wigness, Philip R Osteen, Jinwei Ye, Xuesu Xiao
Dynamics Modeling using Visual Terrain Features for High-Speed Autonomous Off-Road Driving Jason Gibson, Anoushka Alavilli, Erica Tevere, Evangelos Theodorou, Patrick Spieler
ADEPT: Adaptive Diffusion Environment for Policy Transfer Sim-to-Real Youwei Yu, Junhong Xu, Lantao Liu
LunarLoc: Segment-Based Global Localization on the Moon Annika Thomas, Robaire Galliath, Aleksander T Garbuz, Luke R Anger, Cormac ONeill, Trevor E Johst, Dami V Thomas, George Lordos, Jonathan P How
MAGIC-VFM: Meta-Learning Adaptation for Ground Interaction Control with Visual Foundation Models Elena Sorina Lupu, Fengze Xie, James A Preiss, Jedidiah Alindogan, Matthew Anderson, Soon-Jo Chung
Wheeled Lab: Modern Sim2Real for Low-Cost, Open-Source Wheeled Robotics Tyler Han, Preet Shah, Sidharth Rajagopal, Yanda Bao, Sanghun Jung, Sidharth Talia, Gabriel Guo, Bryan Xu, Bhaumik Mehta, Emma Romig, Rosario Scalise, Byron Boots
BEV-Patch-PF: Particle Filtering with BEV-Aerial Feature Matching for Off-Road Geo-Localization Dongmyeong Lee, Jesse Quattrociocchi, Christian Ellis, Rwik Rana, Amanda Adkins, Adam Uccello, Garrett Warnell, Joydeep Biswas
Learning to Predict Mobile Robot Stability in Off-Road Environments Nathaniel Rose, Arif Ahmed, Emanuel Gutierrez-Cornejo, Parikshit Maini
Topics Covered
- Multi-modal perception
- Self-supervised learning
- Sim-to-real transfer
- Sensor fusion techniques (modalities such as radar, thermal, and event camera).
- Learning-based methods for off-road navigation
- Online adaptation
- Traversability estimation
- Foundation models for navigation
- Heterogeneous collaborative robots
- Data collection and generation techniques for off-road environments
- Off-road driving datasets
- Cross-domain transfer for off-road applications
- Scene understanding and reconstruction (e.g. but not limited to, NeRF, Neural Representations)
- Open-set visual recognition methods under constrained conditions (image and video classification, object detection, and segmentation)
- Safe robot operation in uncertain and dynamic environments
- Planning without prior knowledge of an environment
- Approaches to off-road driving that work on multiple types of robots
- Long-term localization and mapping in complex/GPS-denied environments
- Representations of terrain that model physical properties and robot-environment interaction
Submission Instructions
- Submissions should follow the RSS 2025 paper format.
- Reviews will be conducted in a double-blind manner; please ensure your paper is properly anonymized.
- While there is no strict page limit, we recommend a length of 4β8 pages, excluding references and supplementary material.
- Papers must be submitted via OpenReview.
- Accepted papers will be presented as posters, with a select few invited for spotlight presentations.
Important Dates
- Submission Start:
Apr 01 2025 12:00AM UTC-0 - Submission Deadline:
May 01 2025 12:00AM UTC-0May 15 2025 AoE - Acceptance Notification:
May 16 2025 12:00AM UTC-0May 29 2025 AoE - Camera-Ready Deadline:
June 01 2025 12:00AM UTC-0June 05 2025 AoE - Workshop Date: Jun 25th 2025
Online Participation
All sessions, including invited talks, paper presentations, and panel discussions, will be live-streamed. Recordings will be made available after the event to accommodate participants in different time zones. Moreover, online attendees will have the opportunity to actively engage in live Q&A sessions via Zoom chat.
Organizers
Sponsors
