mlsanityData to 14 Oct 2026

Robust Point Cloud Reinforcement Learning via PCA-Based Canonicalization

arXiv 2510.20974 ↗First posted 23 Oct 2025cs.RO cs.LGcited by 1

On the map: Deep reinforcement learning methods → RL for robot locomotion and manipulation

Abstract (arXiv)

Reinforcement Learning (RL) from raw visual input has achieved impressive successes in recent years, yet it remains fragile to out-of-distribution variations such as changes in lighting, color, and viewpoint. Point Cloud Reinforcement Learning (PC-RL) offers a promising alternative by mitigating appearance-based brittleness, but its sensitivity to camera pose mismatches continues to undermine reliability in realistic settings. To address this challenge, we propose PCA Point Cloud (PPC), a canonicalization framework specifically tailored for downstream robotic control. PPC maps point clouds under arbitrary rigid-body transformations to a unique canonical pose, aligning observations to a consistent frame, thereby substantially decreasing viewpoint-induced inconsistencies. In our experiments, we show that PPC improves robustness to unseen camera poses across challenging robotic tasks, providing a principled alternative to domain randomization.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods. Builds on 31 of its 40 references that are in the map. First posted on arXiv 23 Oct 2025.

Builds on

31 of 40 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024PointPatchRL -- Masked Reconstruction Improves Reinforcement Learning on Point Clouds24 Oct

Other references · 30

  1. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  2. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  3. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,838
  4. 2017Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real Worldcited by 1,498
  5. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,264
  6. 2018Soft Actor-Critic Algorithms and Applicationscited by 1,240
  7. 2021Point-BERT: Pre-training 3D Point Cloud Transformers with Masked Point Modelingcited by 653
  8. 2023ManiSkill2: A Unified Benchmark for Generalizable Manipulation Skillscited by 327
  9. 2021Escaping the Big Data Paradigm with Compact Transformerscited by 264
  10. 2017Asymmetric Actor Critic for Image-Based Robot Learningcited by 238
  11. 2019Active Domain Randomizationcited by 170
  12. 2021A System for General In-Hand Object Re-Orientationcited by 165
18 more
  1. 2023PointGPT: Auto-regressively Generative Pre-training from Point Cloudscited by 115
  2. 2022DexPoint: Generalizable Point Cloud Reinforcement Learning for Sim-to-Real Dexterous Manipulationcited by 84
  3. 2019Rotation Invariant Convolutions for 3D Point Clouds Deep Learningcited by 80
  4. 2022Learning Generalizable Dexterous Manipulation from Human Grasp Affordancecited by 51
  5. 2021Generalization in Dexterous Manipulation via Geometry-Aware Multi-Task Learningcited by 50
  6. 2020Rotation-Invariant Local-to-Global Representation Learning for 3D Point Cloudcited by 38
  7. 2022Frame Mining: a Free Lunch for Learning Robotic Manipulation from 3D Point Cloudscited by 32
  8. 2024Demonstrating Agile Flight from Pixels without State Estimationcited by 31
  9. 2022RIConv++: Effective Rotation Invariant Convolutions for 3D Point Clouds Deep Learningcited by 29
  10. 2023RL-ViGen: A Reinforcement Learning Benchmark for Visual Generalizationcited by 28
  11. 2023LapGym -- An Open Source Framework for Reinforcement Learning in Robot-Assisted Laparoscopic Surgerycited by 14
  12. 2024Point Cloud Models Improve Visual Robustness in Robotic Learnerscited by 14
  13. 2020Triangle-Net: Towards Robustness in Point Cloud Learningcited by 13
  14. 2023On the Efficacy of 3D Point Cloud Reinforcement Learningcited by 11
  15. 2023Adversarial Imitation Learning from Visual Observations using Latent Informationcited by 11
  16. 2024Visually Robust Adversarial Imitation Learning from Videos with Contrastive Learningcited by 9
  17. 2024A general framework for rotation invariant point cloud analysiscited by 2
  18. 2025Beyond Domain Randomization: Event-Inspired Perception for Visually Robust Adversarial Imitation from Videoscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026On Preserving Geometrical Invariance for Superpixel Image Classification using Graph Transformercited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Canonical Policy: Learning Canonical 3D Representation for SE(3)-Equivariant Policy
  2. 2026EquiForm: Noise-Robust SE(3)-Equivariant Policy Learning from 3D Point Clouds
  3. 2025PointMapPolicy: Structured Point Cloud Processing for Multi-Modal Imitation Learning
  4. 2025ManiVID-3D: Generalizable View-Invariant Reinforcement Learning for Robotic Manipulation via Disentangled 3D Representations
  5. 2021ReAgent: Point Cloud Registration using Imitation and Reinforcement Learning
  6. 2026PCQA-R1: Advancing Generalized 3D Point Cloud Quality Assessment with Reinforcement Learning
  7. 2025PRISM: Pointcloud Reintegrated Inference via Segmentation and Cross-attention for Manipulation
  8. 2025Towards Fusing Point Cloud and Visual Representations for Imitation Learning