Efficient Skill Acquisition for Complex Manipulation Tasks in Obstructed Environments
Not on the map. Why?
Abstract (arXiv)
Data efficiency in robotic skill acquisition is crucial for operating robots in varied small-batch assembly settings. To operate in such environments, robots must have robust obstacle avoidance and versatile goal conditioning acquired from only a few simple demonstrations. Existing approaches, however, fall short of these requirements. Deep reinforcement learning (RL) enables a robot to learn complex manipulation tasks but is often limited to small task spaces in the real world due to sample inefficiency and safety concerns. Motion planning (MP) can generate collision-free paths in obstructed environments, but cannot solve complex manipulation tasks and requires goal states often specified by a user or object-specific pose estimator. In this work, we propose a system for efficient skill acquisition that leverages an object-centric generative model (OCGM) for versatile goal identification to specify a goal for MP combined with RL to solve complex manipulation tasks in obstructed environments. Specifically, OCGM enables one-shot target object identification and re-identification in new scenes, allowing MP to guide the robot to the target object while avoiding obstacles. This is combined with a skill transition network, which bridges the gap between terminal states of MP and feasible start states of a sample-efficient RL policy. The experiments demonstrate that our OCGM-based one-shot goal identification provides competitive accuracy to other baseline approaches and that our modular framework outperforms competitive baselines, including a state-of-the-art RL algorithm, by a significant margin for complex manipulation tasks in obstructed environments.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 28 of its 47 references that are in the map. First posted on arXiv 6 Mar 2023.
Builds on
28 of 47 references.
References · 28
- 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,635
- 2018Unsupervised Feature Learning via Non-Parametric Instance-level Discriminationcited by 1,838
- 2017Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real Worldcited by 1,401
- 2016End to End Learning for Self-Driving Carscited by 1,000
- 2020Object-Centric Learning with Slot Attentioncited by 865
- 2015End-to-End Training of Deep Visuomotor Policiescited by 847
- 2020CURL: Contrastive Unsupervised Representations for Reinforcement Learningcited by 745
- 2018QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulationcited by 730
- 2019RLBench: The Robot Learning Benchmark & Learning Environmentcited by 669
- 2020Reinforcement Learning with Augmented Datacited by 452
- 2016Deep Reinforcement Learning for Robotic Manipulation with Asynchronous Off-Policy Updatescited by 342
- 2017Leveraging Demonstrations for Deep Reinforcement Learning on Robotics Problems with Sparse Rewardscited by 302
16 more
- 2017Deep Imitation Learning for Complex Manipulation Tasks from Virtual Reality Teleoperationcited by 286
- 2020SPACE: Unsupervised Object-Oriented Scene Representation via Spatial Attention and Decompositioncited by 190
- 2018Making Sense of Vision and Touch: Self-Supervised Learning of Multimodal Representations for Contact-Rich Taskscited by 150
- 2021Coarse-to-Fine Imitation Learning: Robot Manipulation from a Single Demonstrationcited by 92
- 2022You Only Demonstrate Once: Category-Level Manipulation from Single Visual Demonstrationcited by 79
- 2020ReLMoGen: Leveraging Motion Generation in Reinforcement Learning for Mobile Manipulationcited by 47
- 2018A Practical Approach to Insertion with Variable Socket Position Using Deep Reinforcement Learningcited by 35
- 2021Robust Multi-Modal Policies for Industrial Assembly via Reinforcement Learning and Demonstrations: A Large-Scale Studycited by 24
- 2020Guided Uncertainty-Aware Policy Optimization: Combining Learning and Model-Based Strategies for Sample-Efficient Policy Learningcited by 22
- 2020Residual Learning from Demonstration: Adapting DMPs for Contact-rich Manipulationcited by 22
- 2021APEX: Unsupervised, Object-Centric Scene Segmentation and Tracking for Robot Manipulationcited by 16
- 2021InsertionNet -- A Scalable Solution for Insertioncited by 15
- 2021Learning to Search in Task and Motion Planning with Streamscited by 9
- 2020Goal-Conditioned End-to-End Visuomotor Control for Versatile Skill Primitivescited by 7
- 2020Uncertainty-aware Contact-safe Model-based Reinforcement Learningcited by 5
- 2022Residual Robot Learning for Object-Centric Probabilistic Movement Primitivescited by 5
Built on by
Citing papers per year: 2024: 1; 2025: 1.
2 papers.