Learning Models as Functionals of Signed-Distance Fields for Manipulation Planning

This work proposes an optimization-based manipulation planning framework where the objectives are learned functionals of signed-distance fields that represent objects in the scene. Most manipulation planning approaches rely on analytical models and carefully chosen abstractions/state-spaces to be effective. A central question is how models can be obtained from data that are not primarily accurate in their predictions, but, more importantly, enable efficient reasoning within a planning framework, while at the same time being closely coupled to perception spaces. We show that representing objects as signed-distance fields not only enables to learn and represent a variety of models with higher accuracy compared to point-cloud and occupancy measure representations, but also that SDF-based models are suitable for optimization-based planning. To demonstrate the versatility of our approach, we learn both kinematic and dynamic models to solve tasks that involve hanging mugs on hooks and pushing objects on a table. We can unify these quite different tasks within one framework, since SDFs are the common object representation. Video: https://youtu.be/ga8Wlkss7co

[1]  Pieter Abbeel,et al.  Combined task and motion planning through an extensible planner-independent interface layer , 2014, 2014 IEEE International Conference on Robotics and Automation (ICRA).

[2]  Mathew Halm,et al.  ContactNets: Learning of Discontinuous Contact Dynamics with Smooth, Implicit Representations , 2020, CoRL.

[3]  Marc Toussaint,et al.  Differentiable Physics and Stable Modes for Tool-Use and Manipulation Planning , 2018, Robotics: Science and Systems.

[4]  Hao Zhang,et al.  Learning Implicit Fields for Generative Shape Modeling , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Yaron Lipman,et al.  SAL: Sign Agnostic Learning of Shapes From Raw Data , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[6]  Jiajun Wu,et al.  DensePhysNet: Learning Dense Physical Object Representations via Multi-step Dynamic Interactions , 2019, Robotics: Science and Systems.

[7]  Peter R. Florence,et al.  Transporter Networks: Rearranging the Visual World for Robotic Manipulation , 2020, CoRL.

[8]  Russ Tedrake,et al.  A direct method for trajectory optimization of rigid bodies through contact , 2014, Int. J. Robotics Res..

[9]  Jiajun Wu,et al.  Learning 3D Dynamic Scene Representations for Robot Manipulation , 2020, CoRL.

[10]  Kris Hauser,et al.  Semi-infinite programming for trajectory optimization with non-convex obstacles , 2021, Int. J. Robotics Res..

[11]  Gaurav S. Sukhatme,et al.  Learning Equality Constraints for Motion Planning on Manifolds , 2020, CoRL.

[12]  Achim Menges,et al.  Robust Task and Motion Planning for Long-Horizon Architectural Construction Planning , 2020, 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[13]  ErlebenKenny,et al.  Local Optimization for Robust Signed Distance Field Collision , 2020, Proc. ACM Comput. Graph. Interact. Tech..

[14]  Jörg Stückler,et al.  Learning to Identify Physical Parameters from Video Using Differentiable Physics , 2020, GCPR.

[15]  Neel Doshi,et al.  Hybrid Differential Dynamic Programming for Planar Manipulation Primitives , 2020, 2020 IEEE International Conference on Robotics and Automation (ICRA).

[16]  Gordon Wetzstein,et al.  MetaSDF: Meta-learning Signed Distance Functions , 2020, NeurIPS.

[17]  Dieter Fox,et al.  Object Rearrangement Using Learned Implicit Collision Functions , 2020, 2021 IEEE International Conference on Robotics and Automation (ICRA).

[18]  Zoran Popovic,et al.  Discovery of complex behaviors through contact-invariant optimization , 2012, ACM Trans. Graph..

[19]  Jung-Su Ha,et al.  Describing Physics For Physical Reasoning: Force-Based Sequential Manipulation Planning , 2020, IEEE Robotics and Automation Letters.

[20]  Dieter Fox,et al.  Sim-to-Real Task Planning and Execution from Perception via Reactivity and Recovery , 2020, ArXiv.

[21]  Richard A. Newcombe,et al.  DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[22]  Dieter Fox,et al.  SE3-nets: Learning rigid body motion using deep neural networks , 2016, 2017 IEEE International Conference on Robotics and Automation (ICRA).

[23]  Leslie Pack Kaelbling,et al.  Hierarchical Planning in the Now , 2010, Bridging the Gap Between Task and Motion Planning.

[24]  Jeannette Bohg,et al.  OmniHang: Learning to Hang Arbitrary Objects using Contact Point Correspondences and Neural Collision Estimation , 2021, 2021 IEEE International Conference on Robotics and Automation (ICRA).

[25]  Jung-Su Ha,et al.  Deep Visual Reasoning: Learning to Predict Action Sequences for Task and Motion Planning from an Initial Scene Image , 2020, Robotics: Science and Systems.

[26]  Roland Siegwart,et al.  Volumetric Grasping Network: Real-time 6 DOF Grasp Detection in Clutter , 2021, ArXiv.

[27]  Arnulph Fuhrmann,et al.  Distance Fields for Rapid Collision Detection in Physically Based Modeling , 2003 .

[28]  Sergey Levine,et al.  Visual Foresight: Model-Based Deep Reinforcement Learning for Vision-Based Robotic Control , 2018, ArXiv.

[29]  Pulkit Agrawal,et al.  Learning to Plan with Pointcloud Affordances for General-Purpose Dexterous Manipulation , 2010 .

[30]  Hao Li,et al.  Learning to Infer Implicit Surfaces without 3D Supervision , 2019, NeurIPS.

[31]  Jung-Su Ha,et al.  Learning Geometric Reasoning and Control for Long-Horizon Tasks from Visual Input , 2021, 2021 IEEE International Conference on Robotics and Automation (ICRA).

[32]  Yuke Zhu,et al.  Synergies Between Affordance and Geometry: 6-DoF Grasp Detection via Implicit Representations , 2021, Robotics: Science and Systems.

[33]  Kris Hauser,et al.  Semi-Infinite Programming with Complementarity Constraints for Pose Optimization with Pervasive Contact , 2021, 2021 IEEE International Conference on Robotics and Automation (ICRA).

[34]  Marc Toussaint,et al.  Learning to solve sequential physical reasoning problems from a scene image , 2021, Int. J. Robotics Res..

[35]  Meire Fortunato,et al.  Learning Mesh-Based Simulation with Graph Networks , 2020, ArXiv.

[36]  Russ Tedrake,et al.  Keypoints into the Future: Self-Supervised Correspondence in Model-Based Reinforcement Learning , 2020, CoRL.

[37]  Thomas Funkhouser,et al.  Local Implicit Grid Representations for 3D Scenes , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[38]  Syn Schmitt,et al.  Active Inverse Model Learning with Error and Reachable Set Estimates , 2019, 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[39]  Sebastian Nowozin,et al.  Occupancy Networks: Learning 3D Reconstruction in Function Space , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[40]  Leslie Pack Kaelbling,et al.  Integrated Task and Motion Planning , 2020, Annu. Rev. Control. Robotics Auton. Syst..

[41]  Alberto Rodriguez,et al.  Reactive Planar Manipulation with Convex Hybrid MPC , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[42]  Tucker Hermans,et al.  Learning Continuous 3D Reconstructions for Geometrically Aware Grasping , 2020, 2020 IEEE International Conference on Robotics and Automation (ICRA).