Learning to Search in Task and Motion Planning With Streams

Task and motion planning problems in robotics typically combine symbolic planning over discrete task variables with motion optimization over continuous state and action variables, resulting in trajectories that satisfy the logical constraints imposed on the task variables. Symbolic planning can scale exponentially with the number of task variables, so recent works such as PDDLStream [1] have focused on optimistic planning with an incrementally growing set of objects and facts until finding a feasible trajectory. However, this set is exhaustively and uniformly expanded in a breadth-first manner, regardless of the geometric structure of the problem at hand, which makes long-horizon reasoning with large numbers of objects prohibitively time-consuming. To address this issue, we propose a geometrically informed symbolic planner that expands the set of objects and facts in a best-first manner, prioritized by a Graph Neural Network based score that is learned from prior search computations. We evaluate our approach on a diverse set of problems and demonstrate an improved ability to plan in large or difficult scenarios. We also apply our algorithm on a 7DOF robotic arm in several block-stacking manipulation tasks.

[1]  Razvan Pascanu,et al.  Relational inductive biases, deep learning, and graph networks , 2018, ArXiv.

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

[3]  Leslie Pack Kaelbling,et al.  Learning Symbolic Operators for Task and Motion Planning , 2021, 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[4]  Subbarao Kambhampati,et al.  Combining Specialized Reasoners and General Purpose Planners: A Case Study , 1991, AAAI.

[5]  Leslie Pack Kaelbling,et al.  Learning compositional models of robot skills for task and motion planning , 2021, Int. J. Robotics Res..

[6]  Leslie Pack Kaelbling,et al.  Hierarchical task and motion planning in the now , 2011, 2011 IEEE International Conference on Robotics and Automation.

[7]  Beomjoon Kim,et al.  Learning value functions with relational state representations for guiding task-and-motion planning , 2019, CoRL.

[8]  Hammad Mazhar,et al.  Transferable Task Execution from Pixels through Deep Planning Domain Learning , 2020, 2020 IEEE International Conference on Robotics and Automation (ICRA).

[9]  Jung-Su Ha,et al.  Deep Visual Heuristics: Learning Feasibility of Mixed-Integer Programs for Manipulation Planning , 2020, 2020 IEEE International Conference on Robotics and Automation (ICRA).

[10]  Jean-Claude Latombe,et al.  Integrating task and PRM motion planning : Dealing with many infeasible motion planning queries , 2009 .

[11]  Robert Bridson,et al.  Fast Poisson disk sampling in arbitrary dimensions , 2007, SIGGRAPH '07.

[12]  Craig A. Knoblock,et al.  PDDL-the planning domain definition language , 1998 .

[13]  Gregory D. Hager,et al.  Sampling-Based Motion and Symbolic Action Planning with geometric and differential constraints , 2010, 2010 IEEE International Conference on Robotics and Automation.

[14]  Swarat Chaudhuri,et al.  An incremental constraint-based framework for task and motion planning , 2018, Int. J. Robotics Res..

[15]  Rachid Alami,et al.  A Hybrid Approach to Intricate Motion, Manipulation and Task Planning , 2009, Int. J. Robotics Res..

[16]  Álvaro Torralba,et al.  Learning How to Ground a Plan - Partial Grounding in Classical Planning , 2019, AAAI.

[17]  Kelsey R. Allen,et al.  Learning constraint-based planning models from demonstrations , 2020, 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[18]  Jeannette Bohg,et al.  Object-Centric Task and Motion Planning in Dynamic Environments , 2020, IEEE Robotics and Automation Letters.

[19]  Yuke Zhu,et al.  Hierarchical Planning for Long-Horizon Manipulation with Geometric and Symbolic Scene Graphs , 2020, 2021 IEEE International Conference on Robotics and Automation (ICRA).

[20]  Lydia E. Kavraki,et al.  Platform-Independent Benchmarks for Task and Motion Planning , 2018, IEEE Robotics and Automation Letters.

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

[22]  Marc Toussaint,et al.  Logic-Geometric Programming: An Optimization-Based Approach to Combined Task and Motion Planning , 2015, IJCAI.

[23]  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.

[24]  Leslie Pack Kaelbling,et al.  PDDLStream: Integrating Symbolic Planners and Blackbox Samplers via Optimistic Adaptive Planning , 2020, ICAPS.

[25]  Leslie Pack Kaelbling,et al.  Planning with Learned Object Importance in Large Problem Instances using Graph Neural Networks , 2020, AAAI.