Language Movement Primitives: Grounding Language Models in Robot Motion
Collab , Dept. of Mechanical Engineering, Virginia Tech, Blacksburg, VA 24061（合作组，机械工程系，弗吉尼亚理工学院，黑斯堡，VA 24061）；TEA Lab , Dept. of Mechanical Engineering, Virginia Tech, Blacksburg, VA 24061（TEA实验室，机械工程系，弗吉尼亚理工学院，黑斯堡，VA 24061）
Yinlong Dai, Benjamin A. Christie, Daniel J. Evans, Dylan P. Losey, Simon Stepputtis
Enabling robots to perform novel manipulation tasks from natural language instructions remains a fundamental challenge in robotics, despite significant progress in generalized problem solving with foundational models. Large vision and language models (VLMs) are capable of processing high-dimensional input data for visual scene and language understanding, as well as decomposing tasks into a sequence of logical steps; however, they struggle to ground those steps in embodied robot motion. On the other hand, robotics foundation models output action commands, but require in-domain fine-tuning or experience before they are able to perform novel tasks successfully. At its core, there still remains the fundamental challenge of connecting abstract task reasoning with low-level motion control. To address this disconnect, we propose Language Movement Primitives (LMPs), a framework that grounds VLM reasoning in Dynamic Movement Primitive (DMP) parameterization. Our key insight is that DMPs provide a small number of interpretable parameters, and VLMs can set these parameters to specify diverse, continuous, and stable trajectories. Put another way: VLMs can reason over free-form natural language task descriptions, and semantically ground their desired motions into DMPs -- bridging the gap between high-level task reasoning and low-level position and velocity control. Building on this combination of VLMs and DMPs, we formulate our LMP pipeline for zero-shot robot manipulation that effectively completes tabletop manipulation problems by generating a sequence of DMP motions. Across 31 real-world manipulation tasks, we show that LMP achieves 65% task success as compared to 35% for the best performing baseline. See videos at our website: https://collab.me.vt.edu/lmp
Parameterized movement primitives have been extensively used for imitation learning of robotic tasks. However, the high-dimensionality of the parameter space hinders the improvement of such primitives in the reinforcement learning (RL) setting, especially for learning with physical robots. In this paper we propose a novel view on handling the demonstrated trajectories for acquiring low-dimensional, non-linear latent dynamics, using mixtures of probabilistic principal component analyzers (MPPCA) on the movements' parameter space. Moreover, we introduce a new contextual off-policy RL algorithm, …
Agile robotics presents a difficult challenge with robots moving at high speeds requiring precise and low-latency sensing and control. Creating agile motion that accomplishes the task at hand while being safe to execute is a key requirement for agile robots to gain human trust. This requires designing new approaches that are flexible and maintain knowledge over world constraints. In this paper, we consider the problem of building a flexible and adaptive controller for a challenging agile mobile manipulation task of hitting ground strokes on a wheelchair tennis robot. We propose and evaluate an…
Probabilistic Movement Primitives (ProMPs) are a widely used representation of movements for human-robot interaction. They also facilitate the factorization of temporal and spatial structure of movements. In this work we investigate a method to temporally align observations so that when learning ProMPs, information in the spatial structure of the observed motion is maximized while maintaining a smooth phase velocity. We apply the method on recordings of hand trajectories in a two-dimensional reaching task. A system for simultaneous recognition of movement and phase is proposed and performance …
University of Southern California（南加州大学）；MPI for Intelligent Systems（马克斯·普朗克智能系统研究所）
Dynamic Movement Primitives have successfully been used to realize imitation learning, trial-and-error learning, reinforce- ment learning, movement recognition and segmentation and control. Because of this they have become a popular represen- tation for motor primitives. In this work, we showcase how DMPs can be reformulated as a probabilistic linear dynamical system with control inputs. Through this probabilistic repre- sentation of DMPs, algorithms such as Kalman filtering and smoothing are directly applicable to perform inference on pro- prioceptive sensor measurements during execution. We …
We present a novel, reusable and task-agnostic primitive for assessing the outcome of a force-interaction robotic skill, useful e.g.\ for applications such as quality control in industrial manufacturing. The proposed method is easily programmed by kinesthetic teaching, and the desired adaptability and reusability are achieved by machine learning models. The primitive records sensory data during both demonstrations and reproductions of a movement. Recordings include the end-effector's Cartesian pose and exerted wrench at each time step. The collected data are then used to train Gaussian Process…
Aristotle University of Thessaloniki（塞萨洛尼基亚里士多德大学）
In this work, a novel Dynamic Movement Primitive (DMP) formulation is proposed which supports reversibility, i.e. backwards reproduction of a learned trajectory. Apart from sharing all favourable properties of the original DMP, decoupling the teaching of position and velocity profiles and bidirectional drivability along the encoded path are also supported. Original DMP have been extensively used for encoding and reproducing a desired motion pattern in several robotic applications. However, they lack reversibility, which is a useful and expedient property that can be leveraged in many scenarios…
In many robot control problems, factors such as stiffness and damping matrices and manipulability ellipsoids are naturally represented as symmetric positive definite (SPD) matrices, which capture the specific geometric characteristics of those factors. Typical learned skill models such as dynamic movement primitives (DMPs) can not, however, be directly employed with quantities expressed as SPD matrices as they are limited to data in Euclidean space. In this paper, we propose a novel and mathematically principled framework that uses Riemannian metrics to reformulate DMPs such that the resulting…
One of the most important challenges in robotics is producing accurate trajectories and controlling their dynamic parameters so that the robots can perform different tasks. The ability to provide such motion control is closely related to how such movements are encoded. Advances on deep learning have had a strong repercussion in the development of novel approaches for Dynamic Movement Primitives. In this work, we survey scientific literature related to Neural Dynamic Movement Primitives, to complement existing surveys on Dynamic Movement Primitives.
In this paper, we present the probably first application of the popular \emph{Dynamic Movement Primitives (DMP)} approach to the domain of soccer-playing humanoid robots. DMPs are known for their ability to imitate previously demonstrated motions as well as to flexibly adapt to unforeseen changes to the desired trajectory with respect to speed and direction. As demonstrated in this paper, this makes them a useful approach for describing kick motions. Furthermore, we present a mathematical motor model that compensates for the NAO robot's motor control delay as well as a novel minor extension to…
Bosch Center for Artificial Intelligence（博世人工智能中心）
Many possible fields of application of robots in real world settings hinge on the ability of robots to grasp objects. As a result, robot grasping has been an active field of research for many years. With our publication we contribute to the endeavor of enabling robots to grasp, with a particular focus on bin picking applications. Bin picking is especially challenging due to the often cluttered and unstructured arrangement of objects and the often limited graspability of objects by simple top down grasps. To tackle these challenges, we propose a fully self-supervised reinforcement learning appr…
Human-robot collaboration is on the rise. Robots need to increasingly improve the efficiency and smoothness with which they assist humans by properly anticipating a human's intention. To do so, prediction models need to increase their accuracy and responsiveness. This work builds on top of Interaction Movement Primitives with phase estimation and re-formulates the framework to use dynamic human-motion observations which constantly update anticipatory motions. The original framework only considers a single fixed-duration static human observation which is used to perform only one anticipatory mo…
Aristotle University of Thessaloniki（塞萨洛尼基亚里士多德大学）；Chalmers University of Technology（查尔姆斯理工大学）
In this work, the problem of human-robot collaborative object transfer to unknown target poses is addressed. The desired pattern of the end-effector pose trajectory to a known target pose is encoded using DMPs (Dynamic Movement Primitives). During transportation of the object to new unknown targets, a DMP-based reference model and an EKF (Extended Kalman Filter) for estimating the target pose and time duration of the human's intended motion is proposed. A stability analysis of the overall scheme is provided. Experiments using a Kuka LWR4+ robot equipped with an ATI sensor at its end-effector v…
Zhejiang University（浙江大学）；ABB Corporate Research Center（ABB企业研究中心）
We aim to develop an efficient programming method for equipping service robots with the skill of performing sign language motions. This paper addresses the problem of transferring complex dual-arm sign language motions characterized by the coordination among arms and hands from human to robot, which is seldom considered in previous studies of motion retargeting techniques. In this paper, we propose a novel motion retargeting method that leverages graph optimization and Dynamic Movement Primitives (DMPs) for this problem. We employ DMPs in a leader-follower manner to parameterize the original t…
Learning from demonstration (LfD) is a powerful learning method to enable a robot to infer how to perform a task given one or more human demonstrations of the desired task. By learning from end-user demonstration rather than requiring that a domain expert manually programming each skill, robots can more readily be applied to a wider range of real-world applications. Writing robots, as one application of LfD, has become a challenging research topic due to the complexity of human handwriting trajectories. In this paper, we introduce a generalized handwriting-learning system for a physical robot …
Volkswagen Machine Learning Research Lab（大众机器学习研究实验室）
Placing robots outside controlled conditions requires versatile movement representations that allow robots to learn new tasks and adapt them to environmental changes. The introduction of obstacles or the placement of additional robots in the workspace, the modification of the joint range due to faults or range-of-motion constraints are typical cases where the adaptation capabilities play a key role for safely performing the robot's task. Probabilistic movement primitives (ProMPs) have been proposed for representing adaptable movement skills, which are modelled as Gaussian distributions over tr…
Shenzhen Institute of Artificial Intelligence and Robotics for Society (AIRS)（深圳人工智能与机器人研究院（AIRS））；Harbin Institute of Technology（哈尔滨工业大学）；The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））
In this paper, we propose a unified whole-body control framework for velocity-controlled mobile collaborative robots which can distribute task motion into the arm and mobile base according to specific task requirements by adjusting weighting factors. Our framework focuses on addressing two challenging issues in whole-body coordination: 1) different dynamic characteristics of the mobile base and the arm; 2) avoidance of violating both safety and configuration constraints. In addition, our controller involves Coupling Dynamic Movement Primitives to enable the essential capabilities for collabora…
Northeastern University（东北大学）；University of California, San Diego（加利福尼亚大学圣迭戈分校）
We present a deep imitation learning framework for robotic bimanual manipulation in a continuous state-action space. A core challenge is to generalize the manipulation skills to objects in different locations. We hypothesize that modeling the relational information in the environment can significantly improve generalization. To achieve this, we propose to (i) decompose the multi-modal dynamics into elemental movement primitives, (ii) parameterize each primitive using a recurrent graph neural network to capture interactions, and (iii) integrate a high-level planner that composes primitives sequ…
It is desirable for future robots to quickly learn new tasks and adapt learned skills to constantly changing environments. To this end, Probabilistic Movement Primitives (ProMPs) have shown to be a promising framework to learn generalizable trajectory generators from distributions over demonstrated trajectories. However, in practical applications that require high precision in the manipulation of objects, the accuracy of ProMPs is often insufficient, in particular when they are learned in cartesian space from external observations and executed with limited controller gains. Therefore, we propo…
A Probabilistic Movement Primitive (ProMP) defines a distribution over trajectories with an associated feedback policy. ProMPs are typically initialized from human demonstrations and achieve task generalization through probabilistic operations. However, there is currently no principled guidance in the literature to determine how many demonstrations a teacher should provide and what constitutes a "good" demonstration for promoting generalization. In this paper, we present an active learning approach to learning a library of ProMPs capable of task generalization over a given space. We utilize un…
University of Novi Sad（诺维萨德大学）；Jožef Stefan Institute（约瑟夫·斯特凡研究所）
Simultaneously achieving low trajectory errors and compliant control \emph{without} explicit models of the task was effectively addressed with Compliant Movement Primitives (CMP). For a single-robot task, this means that it is accurately following its trajectory, but also exhibits compliant behavior in case of perturbations. In this paper we extend this kind of behavior without explicit models to bimanual tasks. In the presence of an external perturbation on any of the robots, they will both move in synchrony in order to maintain their relative posture, and thus not exert force on the object t…
Agile Robots AG（敏捷机器人公司）；Fraunhofer IPK（弗劳恩霍夫生产设备与设计技术研究所）；Universität Hamburg（汉堡大学）；Intel Asia-Pacific Research & Development Ltd.（英特尔亚太研发有限公司）
Dynamic movement primitives (DMPs) allow complex position trajectories to be efficiently demonstrated to a robot. In contact-rich tasks, where position trajectories alone may not be safe or robust over variation in contact geometry, DMPs have been extended to include force trajectories. However, different task phases or degrees of freedom may require the tracking of either position or force -- e.g., once contact is made, it may be more important to track the force demonstration trajectory in the contact direction. The robot impedance balances between following a position or force reference tra…
Obstacle avoidance for DMPs is still a challenging problem. In our previous work, we proposed a framework for obstacle avoidance based on superquadric potential functions to represent volumes. In this work, we extend our previous work to include the velocity of the trajectory in the definition of the potential. Our formulations guarantee smoother behavior with respect to state-of-the-art point-like methods. Moreover, our new formulation allows to obtain a smoother behavior in proximity of the obstacle than when using a static (i.e. velocity independent) potential. We validate our framework for…
Interaction-Stiffness-Guided Basis Allocation in Dynamic Movement Primitives for Efficient Skill Transfer
Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences（中国科学院宁波材料技术与工程研究所）；University of Chinese Academy of Sciences（中国科学院大学）；The Hong Kong Polytechnic University（香港理工大学）
Brown University（布朗大学）；Mitsubishi Electric Research Labs（三菱电机研究实验室）
Dynamic movement primitives are widely used for learning skills which can be demonstrated to a robot by a skilled human or controller. While their generalization capabilities and simple formulation make them very appealing to use, they possess no strong guarantees to satisfy operational safety constraints for a task. In this paper, we present constrained dynamic movement primitives (CDMP) which can allow for constraint satisfaction in the robot workspace. We present a formulation of a non-linear optimization to perturb the DMP forcing weights regressed by locally-weighted regression to admit a…
Technische Universität Darmstadt（达姆施塔特工业大学）；Max Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）
Movement primitives are an important policy class for real-world robotics. However, the high dimensionality of their parametrization makes the policy optimization expensive both in terms of samples and computation. Enabling an efficient representation of movement primitives facilitates the application of machine learning techniques such as reinforcement on robotics. Motions, especially in highly redundant kinematic structures, exhibit high correlation in the configuration space. For these reasons, prior work has mainly focused on the application of dimensionality reduction techniques in the co…
Movement primitives have the property to accommodate changes in the robot state while maintaining attraction to the original policy. As such, we investigate the use of primitives as a blending mechanism by considering that state deviations from the original policy are caused by user inputs. As the primitive recovers from the user input, it implicitly blends human and robot policies without requiring their weightings -- referred to as arbitration. In this paper, we adopt Dynamical Movement Primitives (DMPs), which allow us to avoid the need for multiple demonstrations, and are fast enough to en…
Imitation learning has been studied widely as a convenient way to transfer human skills to robots. This learning approach is aimed at extracting relevant motion patterns from human demonstrations and subsequently applying these patterns to different situations. Despite many advancements have been achieved, the solutions for coping with unpredicted situations (e.g., obstacles and external perturbations) and high-dimensional inputs are still largely open. In this paper, we propose a novel kernelized movement primitive (KMP), which allows the robot to adapt the learned motor skills and fulfill a …
Western Norway University of Applied Sciences（西挪威应用科学大学）；University of Plymouth（普利茅斯大学）
For a successful deployment of physical Human-Robot Cooperation (pHRC), humans need to be able to teach robots new motor skills quickly. Probabilistic movement primitives (ProMPs) are a promising method to encode a robot's motor skills learned from human demonstrations in pHRC settings. However, most algorithms to learn ProMPs from human demonstrations operate in batch mode, which is not ideal in pHRC. In this paper we propose a new learning algorithm to learn ProMPs incrementally in pHRC settings. Our algorithm incorporates new demonstrations sequentially as they arrive, allowing humans to ob…
Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院）；Bosch Center for Artificial Intelligence（博世人工智能中心）；University of Tübingen（图宾根大学）
Movement Primitives (MPs) are a well-known concept to represent and generate modular trajectories. MPs can be broadly categorized into two types: (a) dynamics-based approaches that generate smooth trajectories from any initial state, e. g., Dynamic Movement Primitives (DMPs), and (b) probabilistic approaches that capture higher-order statistics of the motion, e. g., Probabilistic Movement Primitives (ProMPs). To date, however, there is no method that unifies both, i. e. that can generate smooth trajectories from an arbitrary initial state while capturing higher-order statistics. In this paper,…
In learning from demonstrations, many generative models of trajectories make simplifying assumptions of independence. Correctness is sacrificed in the name of tractability and speed of the learning phase. The ignored dependencies, which often are the kinematic and dynamic constraints of the system, are then only restored when synthesizing the motion, which introduces possibly heavy distortions. In this work, we propose to use those approximate trajectory distributions as close-to-optimal discriminators in the popular generative adversarial framework to stabilize and accelerate the learning pro…
Max Planck Institute（马克斯·普朗克研究所）；TU-Darmstadt（达姆施塔特工业大学）
Probabilistic representations of movement primitives open important new possibilities for machine learning in robotics. These representations are able to capture the variability of the demonstrations from a teacher as a probability distribution over trajectories, providing a sensible region of exploration and the ability to adapt to changes in the robot environment. However, to be able to capture variability and correlations between different joints, a probabilistic movement primitive requires the estimation of a larger number of parameters compared to their deterministic counterparts, that fo…
Bosch Center for Artificial Intelligence（博世人工智能中心）；University of Leoben（莱奥本大学）
Learning complex robot motions necessarily demands to have models that are able to encode and retrieve full-pose trajectories when tasks are defined in operational spaces. Probabilistic movement primitives (ProMPs) stand out as a principled approach that models trajectory distributions learned from demonstrations. ProMPs allow for trajectory modulation and blending to achieve better generalization to novel situations. However, when ProMPs are employed in operational space, their original formulation does not directly apply to full-pose movements including rotational trajectories described by q…
