K-VIL: Keypoints-based Visual Imitation Learning

Visual imitation learning provides efficient and intuitive solutions for robotic systems to acquire novel manipulation skills. However, simultaneously learning geometric task constraints and control policies from visual inputs alone remains a challenging problem. In this paper, we propose an approach for keypoint-based visual imitation (K-VIL) that automatically extracts sparse, object-centric, and embodiment-independent task representations from a small number of human demonstration videos. The task representation is composed of keypoint-based geometric constraints on principal manifolds, their associated local frames, and the movement primitives that are then needed for the task execution. Our approach is capable of extracting such task representations from a single demonstration video, and of incrementally updating them when new demonstrations become available. To reproduce manipulation skills using the learned set of prioritized geometric constraints in novel scenes, we introduce a novel keypoint-based admittance controller. We evaluate our approach in several real-world applications, showcasing its ability to deal with cluttered scenes, new instances of categorical objects, and large object pose and shape variations, as well as its efficiency and robustness in both one-shot and few-shot imitation learning settings. Videos and source code are available at https://sites.google.com/view/k-vil.

SpeedFolding: Learning Efficient Bimanual Folding of Garments

Folding garments reliably and efficiently is a long standing challenge in robotic manipulation due to the complex dynamics and high dimensional configuration space of garments. An intuitive approach is to initially manipulate the garment to a canonical smooth configuration before folding. In this work, we develop SpeedFolding, a reliable and efficient bimanual system, which given user-defined instructions as folding lines, manipulates an initially crumpled garment to (1) a smoothed and (2) a folded configuration. Our primary contribution is a novel neural network architecture that is able to predict pairs of gripper poses to parameterize a diverse set of bimanual action primitives. After learning from 4300 human-annotated and self-supervised actions, the robot is able to fold garments from a random initial configuration in under 120s on average with a success rate of 93%. Real-world experiments show that the system is able to generalize to unseen garments of different color, shape, and stiffness. While prior work achieved 3-6 Folds Per Hour (FPH), SpeedFolding achieves 30-40 FPH.

A Riemannian Take on Human Motion Analysis and Retargeting

Dynamic motions of humans and robots are widely driven by posture-dependent nonlinear interactions between their degrees of freedom. However, these dynamical effects remain mostly overlooked when studying the mechanisms of human movement generation. Inspired by recent works, we hypothesize that human motions are planned as sequences of geodesic synergies, and thus correspond to coordinated joint movements achieved with piecewise minimum energy. The underlying computational model is built on Riemannian geometry to account for the inertial characteristics of the body. Through the analysis of various human arm motions, we find that our model segments motions into geodesic synergies, and successfully predicts observed arm postures, hand trajectories, as well as their respective velocity profiles. Moreover, we show that our analysis can further be exploited to transfer arm motions to robots by reproducing individual human synergies as geodesic paths in the robot configuration space.

Deep Learning Approaches to Grasp Synthesis: A Review

Grasping is the process of picking an object by applying forces and torques at a set of contacts. Recent advances in deep-learning methods have allowed rapid progress in robotic object grasping. We systematically surveyed the publications over the last decade, with a particular interest in grasping an object using all 6 degrees of freedom of the end-effector pose. Our review found four common methodologies for robotic grasping: sampling-based approaches, direct regression, reinforcement learning, and exemplar approaches. Furthermore, we found two 'supporting methods' around grasping that use deep-learning to support the grasping process, shape approximation, and affordances. We have distilled the publications found in this systematic review (85 papers) into ten key takeaways we consider crucial for future robotic grasping and manipulation research. An online version of the survey is available at https://rhys-newbury.github.io/projects/6dof/

Conceptual Design of the Memory System of the Robot Cognitive Architecture ArmarX

We consider the memory system as a key component of any technical cognitive system that can play a central role in bridging the gap between high-level symbolic discrete representations used for reasoning, planning and semantic scene understanding and low-level sensorimotor continuous representations used for control. In this work we described conceptual and technical characteristics such a memory system has to fulfill, together with the underlying data representation. We identify these characteristics based on the experience we gained in developing our ARMAR humanoid robot systems and discuss practical examples that demonstrate what a memory system of a humanoid robot performing tasks in human-centered environments should support, such as multi-modality, introspectability, hetero associativity, predictability or an inherently episodic structure. Based on these characteristics, we extended our robot software framework ArmarX into a unified cognitive architecture that is used in robots of the ARMAR humanoid robot family. Further, we describe, how the development of robot software led us to this novel memory-enabled cognitive architecture and we show how the memory is used by the robots to implement memory-driven behaviors.

Learning to Sequence and Blend Robot Skills via Differentiable Optimization

In contrast to humans and animals who naturally execute seamless motions, learning and smoothly executing sequences of actions remains a challenge in robotics. This paper introduces a novel skill-agnostic framework that learns to sequence and blend skills based on differentiable optimization. Our approach encodes sequences of previously-defined skills as quadratic programs (QP), whose parameters determine the relative importance of skills along the task. Seamless skill sequences are then learned from demonstrations by exploiting differentiable optimization layers and a tailored loss formulated from the QP optimality conditions. Via the use of differentiable optimization, our work offers novel perspectives on multitask control. We validate our approach in a pick-and-place scenario with planar robots, a pouring experiment with a real humanoid robot, and a bimanual sweeping task with a human model.

Geometry-aware Bayesian Optimization in Robotics using Riemannian Matérn Kernels

Bayesian optimization is a data-efficient technique which can be used for control parameter tuning, parametric policy adaptation, and structure design in robotics. Many of these problems require optimization of functions defined on non-Euclidean domains like spheres, rotation groups, or spaces of positive-definite matrices. To do so, one must place a Gaussian process prior, or equivalently define a kernel, on the space of interest. Effective kernels typically reflect the geometry of the spaces they are defined on, but designing them is generally non-trivial. Recent work on the Riemannian Mat\'ern kernels, based on stochastic partial differential equations and spectral theory of the Laplace-Beltrami operator, offers promising avenues towards constructing such geometry-aware kernels. In this paper, we study techniques for implementing these kernels on manifolds of interest in robotics, demonstrate their performance on a set of artificial benchmark functions, and illustrate geometry-aware Bayesian optimization for a variety of robotic applications, covering orientation control, manipulability optimization, and motion planning, while showing its improved performance.

Graph-based Task-specific Prediction Models for Interactions between Deformable and Rigid Objects

Capturing scene dynamics and predicting the future scene state is challenging but essential for robotic manipulation tasks, especially when the scene contains both rigid and deformable objects. In this work, we contribute a simulation environment and generate a novel dataset for task-specific manipulation, involving interactions between rigid objects and a deformable bag. The dataset incorporates a rich variety of scenarios including different object sizes, object numbers and manipulation actions. We approach dynamics learning by proposing an object-centric graph representation and two modules which are Active Prediction Module (APM) and Position Prediction Module (PPM) based on graph neural networks with an encode-process-decode architecture. At the inference stage, we build a two-stage model based on the learned modules for single time step prediction. We combine modules with different prediction horizons into a mixed-horizon model which addresses long-term prediction. In an ablation study, we show the benefits of the two-stage model for single time step prediction and the effectiveness of the mixed-horizon model for long-term prediction tasks. Supplementary material is available at https://github.com/wengzehang/deformable_rigid_interaction_prediction

Learning to Shift Attention for Motion Generation

One challenge of motion generation using robot learning from demonstration techniques is that human demonstrations follow a distribution with multiple modes for one task query. Previous approaches fail to capture all modes or tend to average modes of the demonstrations and thus generate invalid trajectories. The other difficulty is the small number of demonstrations that cannot cover the entire working space. To overcome this problem, a motion generation model with extrapolation ability is needed. Previous works restrict task queries as local frames and learn representations in local frames. We propose a model to solve both problems. For multiple modes, we suggest to learn local latent representations of motion trajectories with a density estimation method based on real-valued non-volume preserving (RealNVP) transformations that provides a set of powerful, stably invertible, and learnable transformations. To improve the extrapolation ability, we propose to shift the attention of the robot from one local frame to another during the task execution. In experiments, we consider the docking problem used also in previous works where a trajectory has to be generated to connect two dockers without collision. We increase complexity of the task and show that the proposed method outperforms other approaches. In addition, we evaluate the approach in real robot experiments.

Object and Relation Centric Representations for Push Effect Prediction

Pushing is an essential non-prehensile manipulation skill used for tasks ranging from pre-grasp manipulation to scene rearrangement, reasoning about object relations in the scene, and thus pushing actions have been widely studied in robotics. The effective use of pushing actions often requires an understanding of the dynamics of the manipulated objects and adaptation to the discrepancies between prediction and reality. For this reason, effect prediction and parameter estimation with pushing actions have been heavily investigated in the literature. However, current approaches are limited because they either model systems with a fixed number of objects or use image-based representations whose outputs are not very interpretable and quickly accumulate errors. In this paper, we propose a graph neural network based framework for effect prediction and parameter estimation of pushing actions by modeling object relations based on contacts or articulations. Our framework is validated both in real and simulated environments containing different shaped multi-part objects connected via different types of joints and objects with different masses. Our approach enables the robot to predict and adapt the effect of a pushing action as it observes the scene. Further, we demonstrate 6D effect prediction in the lever-up action in the context of robot-based hard-disk disassembly.