2000 character limit reached
Graph-based 3D Collision-distance Estimation Network with Probabilistic Graph Rewiring (2310.04044v2)
Published 6 Oct 2023 in cs.RO
Abstract: We aim to solve the problem of data-driven collision-distance estimation given 3-dimensional (3D) geometries. Conventional algorithms suffer from low accuracy due to their reliance on limited representations, such as point clouds. In contrast, our previous graph-based model, GraphDistNet, achieves high accuracy using edge information but incurs higher message-passing costs with growing graph size, limiting its applicability to 3D geometries. To overcome these challenges, we propose GDN-R, a novel 3D graph-based estimation network.GDN-R employs a layer-wise probabilistic graph-rewiring algorithm leveraging the differentiable Gumbel-top-K relaxation. Our method accurately infers minimum distances through iterative graph rewiring and updating relevant embeddings. The probabilistic rewiring enables fast and robust embedding with respect to unforeseen categories of geometries. Through 41,412 random benchmark tasks with 150 pairs of 3D objects, we show GDN-R outperforms state-of-the-art baseline methods in terms of accuracy and generalizability. We also show that the proposed rewiring improves the update performance reducing the size of the estimation model. We finally show its batch prediction and auto-differentiation capabilities for trajectory optimization in both simulated and real-world scenarios.
- J. Vorndamme, M. Schappler, and S. Haddadin, “Collision detection, isolation and identification for humanoids,” in Proc. Int’l Conf. on Robotics and Automation, pp. 4754–4761, IEEE, 2017.
- E. G. Gilbert, D. W. Johnson, and S. S. Keerthi, “A fast procedure for computing the distance between complex objects in three-dimensional space,” Journal on Robotics and Automation, vol. 4, no. 2, pp. 193–203, 1988.
- J. Chase Kew, B. Ichter, M. Bandari, T.-W. E. Lee, and A. Faust, “Neural collision clearance estimator for batched motion planning,” in Proc. Int’l Workshop on Algorithmic Foundations of Robotics, pp. 73–89, 2020.
- J. Kim and F. C. Park, “Pairwisenet: Pairwise collision distance learning for high-dof robot systems,” in Proc. Conf. on robot learning, vol. 229, pp. 2863–2877, 2023.
- Y. Zhi, N. Das, and M. Yip, “Diffco: Autodifferentiable proxy collision detection with multiclass labels for safety-aware trajectory optimization,” Trans. on Robotics, vol. 38, no. 5, pp. 2668–2685, 2022.
- M. Koptev, N. Figueroa, and A. Billard, “Neural joint space implicit signed distance functions for reactive robot manipulator control,” IEEE Robotics and Automation Letters, vol. 8, pp. 480–487, 2022.
- J. Michaux, Q. Chen, Y. Kwon, and R. Vasudevan, “Reachability-based trajectory design with neural implicit safety constraints,” Proc. Robotics: Science and Systems, 2023.
- Y. Kim, J. Kim, and D. Park, “GraphDistNet: A graph-based collision-distance estimator for gradient-based trajectory optimization,” IEEE Robotics and Automation Letters, vol. 7, pp. 11118–11125, 2022.
- W. Kool, H. Van Hoof, and M. Welling, “Stochastic beams and where to find them: The gumbel-top-k trick for sampling sequences without replacement,” in Proc. Int’l Conf. on Machine Learning, pp. 3499–3508, PMLR, 2019.
- S. M. Xie and S. Ermon, “Reparameterizable subset sampling via continuous relaxations,” Proc. Int’l Joint Conf. on Artificial Intelligence, 2019.
- J. Gilmer, S. S. Schoenholz, P. F. Riley, O. Vinyals, and G. E. Dahl, “Neural message passing for quantum chemistry,” in Proc. Int’l Conf. on Machine Learning, vol. 70, pp. 1263–1272, PMLR, 2017.
- Y. You, L. Shao, T. Migimatsu, and J. Bohg, “Omnihang: Learning to hang arbitrary objects using contact point correspondences and neural collision estimation,” in Proc. Int’l Conf. on Robotics and Automation, pp. 5921–5927, IEEE, 2021.
- R. G. Luque, J. L. Comba, and C. M. Freitas, “Broad-phase collision detection using semi-adjusting bsp-trees,” in Proceedings of the symposium on Interactive 3D graphics and games, pp. 179–186, ACM New York, NY, USA, 2005.
- S. Quinlan, “Efficient distance computation between non-convex objects,” in Proc. Int’l Conf. on Robotics and Automation, pp. 3324–3329, IEEE, 1994.
- T. Larsson and T. Akenine-Möller, “A dynamic bounding volume hierarchy for generalized collision detection,” Computers & Graphics, vol. 30, no. 3, pp. 450–459, 2006.
- M. Benallegue, A. Escande, S. Miossec, and A. Kheddar, “Fast c 1 proximity queries using support mapping of sphere-torus-patches bounding volumes,” in Proc. Int’l Conf. on Robotics and Automation, pp. 483–488, IEEE, 2009.
- J. Schulman, Y. Duan, J. Ho, A. Lee, I. Awwal, H. Bradlow, J. Pan, S. Patil, K. Goldberg, and P. Abbeel, “Motion planning with sequential convex optimization and convex collision checking,” Int’l J. of Robotics Research, vol. 33, no. 9, pp. 1251–1270, 2014.
- O. S. Lawlor and L. V. Kalée, “A voxel-based parallel collision detection algorithm,” in Proceedings of the 16th International Conference on Supercomputing, ACM New York, NY, USA, 2002.
- N. Das, N. Gupta, and M. Yip, “Fastron: An online learning-based model and active learning strategy for proxy collision detection,” in Proc. Conf. on robot learning, pp. 496–504, PMLR, 2017.
- J. Muñoz, P. Lehner, L. E. Moreno, A. Albu-Schäffer, and M. A. Roa, “Collisiongp: Gaussian process-based collision checking for robot motion planning,” IEEE Robotics and Automation Letters, vol. 8, pp. 4036–4043, 2023.
- N. Das and M. C. Yip, “Forward kinematics kernel for improved proxy collision checking,” IEEE Robotics and Automation Letters, vol. 5, pp. 2349–2356, 2020.
- D. Son and B. Kim, “Local object crop collision network for efficient simulation of non-convex objects in gpu-based simulators,” Proc. Robotics: Science and Systems, 2023.
- M. Danielczuk, A. Mousavian, C. Eppner, and D. Fox, “Object rearrangement using learned implicit collision functions,” in Proc. Int’l Conf. on Robotics and Automation, pp. 6010–6017, IEEE, 2021.
- M. Yoon, M. Kang, D. Park, and S.-E. Yoon, “Learning-based initialization of trajectory optimization for path-following problems of redundant manipulators,” in Proc. Int’l Conf. on Robotics and Automation, pp. 9686–9692, IEEE, 2023.
- M. Zucker, N. Ratliff, A. D. Dragan, M. Pivtoraiko, M. Klingensmith, C. M. Dellin, J. A. Bagnell, and S. S. Srinivasa, “Chomp: Covariant hamiltonian optimization for motion planning,” Int’l J. of Robotics Research, vol. 32, no. 9-10, pp. 1164–1193, 2013.
- M. Fey and J. E. Lenssen, “Fast graph representation learning with PyTorch Geometric,” in ICLR Workshop on Representation Learning on Graphs and Manifolds, 2019.
- C. J. Maddison, D. Tarlow, and T. Minka, “A* sampling,” Conf. on Neural Information Processing Systems, vol. 27, 2014.
- C. J. Maddison, A. Mnih, and Y. W. Teh, “The concrete distribution: A continuous relaxation of discrete random variables,” Proc. Int’l. Conf. on Learning Representations, 2016.
- M. Garland and P. S. Heckbert, “Surface simplification using quadric error metrics,” in Proceedings of the annual conference on Computer graphics and interactive techniques, pp. 209–216, 1997.
- S. LaValle, “Rapidly-exploring random trees: A new tool for path planning,” Research Report 9811, 1998.
- C. R. Qi, L. Yi, H. Su, and L. J. Guibas, “Pointnet++: Deep hierarchical feature learning on point sets in a metric space,” Conf. on Neural Information Processing Systems, vol. 30, 2017.
- A. Murali, A. Mousavian, C. Eppner, A. Fishman, and D. Fox, “Cabinet: Scaling neural collision detection for object rearrangement with procedural scene generation,” in Proc. Int’l Conf. on Robotics and Automation, pp. 1866–1874, 2023.
- J. Pan, S. Chitta, and D. Manocha, “Fcl: A general purpose library for collision and proximity queries,” in Proc. Int’l Conf. on Robotics and Automation, pp. 3859–3866, IEEE, 2012.
Sponsor
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.