Long-horizon tasks, usually characterized by complex subtask dependencies, present a significant challenge in manipulation planning. Skill chaining is a practical approach to solving unseen tasks by combining learned skill priors. However, such methods are myopic if sequenced greedily and face scalability issues with search-based planning strategy. To address these challenges, we introduce Generative Skill Chaining~(GSC), a probabilistic framework that learns skill-centric diffusion models and composes their learned distributions to generate long-horizon plans during inference. GSC samples from all skill models in parallel to efficiently solve unseen tasks while enforcing geometric constraints. We evaluate the method on various long-horizon tasks and demonstrate its capability in reasoning about action dependencies, constraint handling, and generalization, along with its ability to replan in the face of perturbations. We show results in simulation and on real robot to validate the efficiency and scalability of GSC, highlighting its potential for advancing long-horizon task planning. More details are available at: https://generative-skill-chaining.github.io/
University of Southern California（南加州大学）；NVIDIA（英伟达）；California Institute of Technology（加州理工学院）；The University of Texas at Austin（德克萨斯大学奥斯汀分校）
Skill chaining is a promising approach for synthesizing complex behaviors by sequentially combining previously learned skills. Yet, a naive composition of skills fails when a policy encounters a starting state never seen during its training. For successful skill chaining, prior approaches attempt to widen the policy's starting state distribution. However, these approaches require larger state distributions to be covered as more policies are sequenced, and thus are limited to short skill sequences. In this paper, we propose to chain multiple policies without excessively large initial state dist…
Robots deployed in many real-world settings need to be able to acquire new skills and solve new tasks over time. Prior works on planning with skills often make assumptions on the structure of skills and tasks, such as subgoal skills, shared skill implementations, or task-specific plan skeletons, which limit adaptation to new skills and tasks. By contrast, we propose doing task planning by jointly searching in the space of parameterized skills using high-level skill effect models learned in simulation. We use an iterative training procedure to efficiently generate relevant data to train such mo…
UC San Diego（加州大学圣地亚哥分校）；Meta AI Research（Meta人工智能研究院）；UC Berkeley（加州大学伯克利分校）
We study a modular approach to tackle long-horizon mobile manipulation tasks for object rearrangement, which decomposes a full task into a sequence of subtasks. To tackle the entire task, prior work chains multiple stationary manipulation skills with a point-goal navigation skill, which are learned individually on subtasks. Although more effective than monolithic end-to-end RL policies, this framework suffers from compounding errors in skill chaining, e.g., navigating to a bad location where a stationary manipulation skill can not reach its target to manipulate. To this end, we propose that th…
DROM: A Language-Guided Diffusion Framework for Multi-Skill Robotic Manipulation
University of Applied Science and Arts of Southern Switzerland (SUPSI)（瑞士南部应用科学与艺术大学）；Institute of Systems and Technologies for Sustainable Production (ISTePS)（可持续生产系统与技术研究所）；Istituto Dalle Molle di studi sull’intelligenza artificiale (IDSIA)（达勒莫尔人工智能研究所）
Uni-Skill: Building Self-Evolving Skill Repository for Generalizable Robotic Manipulation
Key Laboratory of AI Safety of CAS（中国科学院人工智能安全重点实验室）；Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）；University of Chinese Academy of Sciences（中国科学院大学）
Google Research（谷歌研究院）；Berkeley AI Research（伯克利人工智能研究院）；UC Berkeley（加州大学伯克利分校）
Reinforcement learning can train policies that effectively perform complex tasks. However for long-horizon tasks, the performance of these methods degrades with horizon, often necessitating reasoning over and chaining lower-level skills. Hierarchical reinforcement learning aims to enable this by providing a bank of low-level skills as action abstractions. Hierarchies can further improve on this by abstracting the space states as well. We posit that a suitable state abstraction should depend on the capabilities of the available lower-level policies. We propose Value Function Spaces: a simple ap…
Recent works have shown how the reasoning capabilities of Large Language Models (LLMs) can be applied to domains beyond natural language processing, such as planning and interaction for robots. These embodied problems require an agent to understand many semantic aspects of the world: the repertoire of skills available, how these skills influence the world, and how changes to the world map back to the language. LLMs planning in embodied environments need to consider not just what skills to do, but also how and when to do them - answers that change over time in response to the agent's own choice…
Planning Method for Skill-Based Control of Robots Using a PLC as Skill Trigger
Fraunhofer Institute for Casting, Composite and Processing Technology IGCV（弗劳恩霍夫铸造、复合材料与加工技术研究所 IGCV）；Institute for Machine Tools and Industrial Management, Technical University of Munich（慕尼黑工业大学机床与工业管理研究所）
We describe an algorithm for motion planning based on expert demonstrations of a skill. In order to teach robots to perform complex object manipulation tasks that can generalize robustly to new environments, we must (1) learn a representation of the effects of a task and (2) find an optimal trajectory that will reproduce these effects in a new environment. We represent robot skills in terms of a probability distribution over features learned from multiple expert demonstrations. When utilizing a skill in a new environment, we compute feature expectations over trajectory samples in order to stoc…
We propose a learning-from-demonstration approach for grounding actions from expert data and an algorithm for using these actions to perform a task in new environments. Our approach is based on an application of sampling-based motion planning to search through the tree of discrete, high-level actions constructed from a symbolic representation of a task. Recursive sampling-based planning is used to explore the space of possible continuous-space instantiations of these actions. We demonstrate the utility of our approach with a magnetic structure assembly task, showing that the robot can intellig…
In this paper, we propose a data-driven skill learning approach to solve highly dynamic manipulation tasks entirely from offline teleoperated play data. We use a bilateral teleoperation system to continuously collect a large set of dexterous and agile manipulation behaviors, which is enabled by providing direct force feedback to the operator. We jointly learn the state conditional latent skill distribution and skill decoder network in the form of goal-conditioned policy and skill conditional state transition dynamics using a two-stage generative modeling framework. This allows one to perform r…
Decompose and Reorganize: Planning with Primitives and Visuomotor Policies Learned from Demonstrations
The University of Hong Kong（香港大学）；JD.com (JingDong)（京东）；Nanyang Technological University (NTU)（南洋理工大学）；Southern University of Science and Technology (SUSTech)（南方科技大学）；Huawei Technologies（华为技术有限公司）；The Chinese University of Hong Kong (CUHK)（香港中文大学）
This paper presents JEDAI, an AI system designed for outreach and educational efforts aimed at non-AI experts. JEDAI features a novel synthesis of research ideas from integrated task and motion planning and explainable AI. JEDAI helps users create high-level, intuitive plans while ensuring that they will be executable by the robot. It also provides users customized explanations about errors and helps improve their understanding of AI planning as well as the limits and capabilities of the underlying robot system.
DeCo: Task Decomposition and Skill Composition for Zero-Shot Generalization in Long-Horizon 3D Manipulation
State Key Laboratory for Novel Software Technology, Nanjing University（南京大学新型软件技术国家重点实验室）；School of Electrical and Electronic Engineering, Nanyang Technological University（南洋理工大学电子与电气工程学院）；School of Computing, National University of Singapore（新加坡国立大学计算机学院）；Huawei Noah’s Ark Lab (AI Lab), China（华为诺亚实验室（AI实验室））
University of Freiburg（弗赖堡大学）；University of Technology Nuremberg（纽伦堡工业大学）
Everyday tasks of long-horizon and comprising a sequence of multiple implicit subtasks still impose a major challenge in offline robot control. While a number of prior methods aimed to address this setting with variants of imitation and offline reinforcement learning, the learned behavior is typically narrow and often struggles to reach configurable long-horizon goals. As both paradigms have complementary strengths and weaknesses, we propose a novel hierarchical approach that combines the strengths of both methods to learn task-agnostic long-horizon policies from high-dimensional camera observ…
The objective of lifelong reinforcement learning (RL) is to optimize agents which can continuously adapt and interact in changing environments. However, current RL approaches fail drastically when environments are non-stationary and interactions are non-episodic. We propose Lifelong Skill Planning (LiSP), an algorithmic framework for non-episodic lifelong RL based on planning in an abstract space of higher-order skills. We learn the skills in an unsupervised manner using intrinsic rewards and plan over the learned skills using a learned dynamics model. Moreover, our framework permits skill dis…
Learning a robot motor skill from scratch is impractically slow; so much so that in practice, learning must be bootstrapped using a good skill policy obtained from human demonstration. However, relying on human demonstration necessarily degrades the autonomy of robots that must learn a wide variety of skills over their operational lifetimes. We propose using kinematic motion planning as a completely autonomous, sample efficient way to bootstrap motor skill learning for object manipulation. We demonstrate the use of motion planners to bootstrap motor skills in two complex object manipulation sc…
Lund University（隆德大学）；University of Patras（帕特雷大学）；Stanford University（斯坦福大学）
In modern industrial settings with small batch sizes it should be easy to set up a robot system for a new task. Strategies exist, e.g. the use of skills, but when it comes to handling forces and torques, these systems often fall short. We introduce an approach that provides a combination of task-level planning with targeted learning of scenario-specific parameters for skill-based systems. We propose the following pipeline: (1) the user provides a task goal in the planning language PDDL, (2) a plan (i.e., a sequence of skills) is generated and the learnable parameters of the skills are automati…
Lund University（隆德大学）；University of Patras（帕特雷大学）；Stanford University（斯坦福大学）
One of today's goals for industrial robot systems is to allow fast and easy provisioning for new tasks. Skill-based systems that use planning and knowledge representation have long been one possible answer to this. However, especially with contact-rich robot tasks that need careful parameter settings, such reasoning techniques can fall short if the required knowledge not adequately modeled. We show an approach that provides a combination of task-level planning and reasoning with targeted learning of skill parameters for a task at hand. Starting from a task goal formulated in PDDL, the learnabl…
We propose a novel approach for planning agents to compose abstract skills via observing and learning from historical interactions with the world. Our framework operates in a Markov state-space model via a set of actions under unknown pre-conditions. We formulate skills as high-level abstract policies that propose action plans based on the current state. Each policy learns new plans by observing the states' transitions while the agent interacts with the world. Such an approach automatically learns new plans to achieve specific intended effects, but the success of such plans is often dependent …
MIT Computer Science and Artificial Intelligence Laboratory（麻省理工学院计算机科学与人工智能实验室）
Decision-making is challenging in robotics environments with continuous object-centric states, continuous actions, long horizons, and sparse feedback. Hierarchical approaches, such as task and motion planning (TAMP), address these challenges by decomposing decision-making into two or more levels of abstraction. In a setting where demonstrations and symbolic predicates are given, prior work has shown how to learn symbolic operators and neural samplers for TAMP with manually designed parameterized policies. Our main contribution is a method for learning parameterized polices in combination with …
