mlsanityData to 14 Oct 2026

Cross-Modal Reinforcement Learning for Navigation with Degraded Depth Measurements

arXiv 2603.22182 ↗First posted 23 Mar 2026cs.ROcited by 0

On the map: Deep reinforcement learning methods → RL for robot navigation and UAV control

Abstract (arXiv)

This paper presents a cross-modal learning framework that exploits complementary information from depth and grayscale images for robust navigation. We introduce a Cross-Modal Wasserstein Autoencoder that learns shared latent representations by enforcing cross-modal consistency, enabling the system to infer depth-relevant features from grayscale observations when depth measurements are corrupted. The learned representations are integrated with a Reinforcement Learning-based policy for collision-free navigation in unstructured environments when depth sensors experience degradation due to adverse conditions such as poor lighting or reflective surfaces. Simulation and real-world experiments demonstrate that our approach maintains robust performance under significant depth degradation and successfully transfers to real environments.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL for robot navigation and UAV control, one of 15 directions in Deep reinforcement learning methods. Builds on 23 of its 35 references that are in the map. First posted on arXiv 23 Mar 2026.

Builds on

23 of 35 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025Enhancing Deep Reinforcement Learning-based Robot Navigation Generalization through Scenario Augmentation3 Mar
  2. 2024Vision-Based Deep Reinforcement Learning of UAV Autonomous Navigation Using Privileged Information9 Dec
  3. 2024Reinforcement Learning for Collision-free Flight Exploiting Deep Collision Encoding6 Feb

Other references · 20

  1. 2020TartanAir: A Dataset to Push the Limits of Visual SLAMcited by 506
  2. 2017Wasserstein Auto-Encoderscited by 373
  3. 2023Multimodal Deep Learningcited by 258
  4. 2015Cross Modal Distillation for Supervision Transfercited by 175
  5. 2017Sparse-to-Dense: Depth Prediction from Sparse Depth Samples and a Single Imagecited by 161
  6. 2021Learning High-Speed Flight in the Wildcited by 128
  7. 2020Sample Factory: Egocentric 3D Control from Pixels at 100000 FPS with Asynchronous Reinforcement Learningcited by 93
  8. 2021Perception and Sensing for Autonomous Vehicles Under Adverse Weather Conditions: A Surveycited by 82
  9. 2019Uncertainty Quantification with Statistical Guarantees in End-to-End Autonomous Driving Controlcited by 61
  10. 2017Learning to Fly by Crashingcited by 41
  11. 2022Learning Perception-Aware Agile Flight in Cluttered Environmentscited by 23
  12. 2022CERBERUS: Autonomous Legged and Aerial Robotic Exploration in the Tunnel and Urban Circuits of the DARPA Subterranean Challengecited by 17
8 more
  1. 2025Aerial Gym Simulator: A Framework for Highly Parallelized Simulation of Aerial Robotscited by 15
  2. 2022Learning an Efficient Multimodal Depth Completion Modelcited by 7
  3. 2020Multi-level Cross-modal Interaction Network for RGB-D Salient Object Detectioncited by 6
  4. 2019Stacked Wasserstein Autoencodercited by 5
  5. 2024Degradation Resilient LiDAR-Radar-Inertial Odometrycited by 2
  6. 2024Robust High-Speed State Estimation for Off-road Navigation using Radar Velocity Factorscited by 2
  7. 2025Cross-modal feature fusion for robust point cloud registration with ambiguous geometrycited by 2
  8. 2019Vision-Depth Landmarks and Inertial Fusion for Navigation in Degraded Visual Environmentscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2023Robust Navigation with Cross-Modal Fusion and Knowledge Transfer
  2. 2019Learning Visuomotor Policies for Aerial Navigation Using Cross-Modal Representations
  3. 2026Cross-Modal Navigation with Multi-Agent Reinforcement Learning
  4. 2025Depth Transfer: Learning to See Like a Simulator for Real-World Drone Navigation
  5. 2025X-Nav: Learning End-to-End Cross-Embodiment Navigation for Mobile Robots
  6. 2018Reinforced Cross-Modal Matching and Self-Supervised Imitation Learning for Vision-Language Navigation
  7. 2026MAMMOTH: A Multi-Modal End-to-End Policy for Off-Road Mobility Robust to Missing Modality ↗
  8. 2026Robust Scene Transfer for PointGoal Navigation via Privileged Sensor Guided Contrastive Learning