Optimistic Reinforcement Learning-Based Skill Insertions for Task and Motion Planning
On the map: Deep reinforcement learning methods → RL for robot locomotion and manipulation
Abstract (arXiv)
Task and motion planning (TAMP) for robotics manipulation necessitates long-horizon reasoning involving versatile actions and skills. While deterministic actions can be crafted by sampling or optimizing with certain constraints, planning actions with uncertainty, i.e., probabilistic actions, remains a challenge for TAMP. On the contrary, Reinforcement Learning (RL) excels in acquiring versatile, yet short-horizon, manipulation skills that are robust with uncertainties. In this letter, we design a method that integrates RL skills into TAMP pipelines. Besides the policy, a RL skill is defined with data-driven logical components that enable the skill to be deployed by symbolic planning. A plan refinement sub-routine is designed to further tackle the inevitable effect uncertainties. In the experiments, we compare our method with baseline hierarchical planning from both TAMP and RL fields and illustrate the strength of the method. The results show that by embedding RL skills, we extend the capability of TAMP to domains with probabilistic skills, and improve the planning efficiency compared to the previous methods.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods. Builds on 18 of its 37 references that are in the map. First posted on arXiv 15 Oct 2025.
Builds on
18 of 37 references.
References · 18
- 2016Learning Hand-Eye Coordination for Robotic Grasping with Deep Learning and Large-Scale Data Collectioncited by 316
- 2018PDDLStream: Integrating Symbolic Planners and Blackbox Samplers via Optimistic Adaptive Planningcited by 150
- 2019Online Replanning in Belief Space for Partially Observable Task and Motion Problemscited by 61
- 2021Learning Symbolic Operators for Task and Motion Planningcited by 58
- 2022Learning Neuro-Symbolic Skills for Bilevel Planningcited by 58
- 2018PEORL: Integrating Symbolic Planning and Hierarchical Reinforcement Learning for Robust Decision-Makingcited by 52
- 2020Deep Visual Reasoning: Learning to Predict Action Sequences for Task and Motion Planning from an Initial Scene Imagecited by 48
- 2020Learning compositional models of robot skills for task and motion planningcited by 48
- 2021Long-Horizon Manipulation of Unknown Objects via Task and Motion Planning with Estimated Affordancescited by 40
- 2019Dot-to-Dot: Explainable Hierarchical Reinforcement Learning for Robotic Manipulationcited by 35
- 2019Object-Centric Task and Motion Planning in Dynamic Environmentscited by 35
- 2021Discovering State and Action Abstractions for Generalized Task and Motion Planningcited by 18
6 more
- 2019Learning STRIPS Action Models with Classical Planningcited by 14
- 2021Efficient learning of goal-oriented push-grasping synergy in cluttercited by 12
- 2021Learning to Search in Task and Motion Planning with Streamscited by 10
- 2023Learning Reusable Manipulation Strategiescited by 9
- 2018Sampling-Based Methods for Factored Task and Motion Planningcited by 8
- 2016Backward-Forward Search for Manipulation Planningcited by 2
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2023Offline Skill Generalization via Task and Motion Planning
- 2024Partially Observable Task and Motion Planning with Uncertainty and Risk Awareness
- 2026ManiDreams: An Open-Source Library for Robust Object Manipulation via Uncertainty-aware Task-specific Intuitive Physics
- 2026Learning Diffusion Policy from Primitive Skills for Robot Manipulation
- 2025Robotic Long-Horizon Manipulation with Bayesian Non-parametric Skill Priors
- 2023Generative Skill Chaining: Long-Horizon Skill Planning with Diffusion Models
- 2025Skill-based Safe Reinforcement Learning with Risk Planning
- 2024Guided Reinforcement Learning for Robust Multi-Contact Loco-Manipulation