Vision-based object detectors are a crucial basis for robotics applications as they provide valuable information about object localization in the environment. These need to ensure high reliability in different lighting conditions, occlusions, and visual artifacts, all while running in real-time. Collecting and annotating real-world data for these networks is prohibitively time consuming and costly, especially for custom assets, such as industrial objects, making it untenable for generalization to in-the-wild scenarios. To this end, we present Synthetica, a method for large-scale synthetic data generation for training robust state estimators. This paper focuses on the task of object detection, an important problem which can serve as the front-end for most state estimation problems, such as pose estimation. Leveraging data from a photorealistic ray-tracing renderer, we scale up data generation, generating 2.7 million images, to train highly accurate real-time detection transformers. We present a collection of rendering randomization and training-time data augmentation techniques conducive to robust sim-to-real performance for vision tasks. We demonstrate state-of-the-art performance on the task of object detection while having detectors that run at 50–100Hz which is 9 times faster than the prior state-of-the-art (SOTA). We further demonstrate the usefulness of our training methodology for robotics applications by showcasing a pipeline for use in the real world with custom objects for which there do not exist prior datasets. Our work highlights the importance of scaling synthetic data generation for robust sim-to-real transfer while achieving the fastest real-time inference speeds. Videos and supplementary information can be found at https://sites.google.com/view/synthetica-vision

Robotics policies are always subjected to complex, second order dynamics that entangle their actions with resulting states. In reinforcement learning (RL) contexts, policies have the burden of deciphering these complicated interactions over massive amounts of experience and complex reward functions to learn how to accomplish tasks. Moreover, policies typically issue actions directly to controllers like Operational Space Control (OSC) or joint PD control, which induces straightline motion towards these action targets in task or joint space. However, straightline motion in these spaces for the most part do not capture the rich, nonlinear behavior our robots need to exhibit, shifting the burden of discovering these behaviors more completely to the agent. Unlike these simpler controllers, geometric fabrics capture a much richer and desirable set of behaviors via artificial, second order dynamics grounded in nonlinear geometry. These artificial dynamics shift the uncontrolled dynamics of a robot via an appropriate control law to form behavioral dynamics. Behavioral dynamics unlock a new action space and safe, guiding behavior over which RL policies are trained. Behavioral dynamics enable bang-bang-like RL policy actions that are still safe for real robots, simplify reward engineering, and help sequence real-world, high-performance policies. We describe the framework more generally and create a specific instantiation for the problem of dexterous, in-hand reorientation of a cube by a highly actuated robot hand.

Recent work has demonstrated the ability of deep reinforcement learning (RL) algorithms to learn complex robotic behaviours in simulation, including in the domain of multi-fingered manipulation. However, such models can be challenging to transfer to the real world due to the gap between simulation and reality. In this paper, we present our techniques to train a) a policy that can perform robust dexterous manipulation on an anthropomorphic robot hand and b) a robust pose estimator suitable for providing reliable real-time information on the state of the object being manipulated. Our policies are trained to adapt to a wide range of conditions in simulation. Consequently, our vision-based policies significantly outperform the best vision policies in the literature on the same reorientation task and are competitive with policies that are given privileged state information via motion capture systems. Our work reaffirms the possibilities of sim-to-real transfer for dexterous manipulation in diverse kinds of hardware and simulator setups, and in our case, with the Allegro Hand and Isaac Gym GPU-based simulation. Furthermore, it opens up possibilities for researchers to achieve such results with commonly-available, affordable robot hands and cameras. Videos of the resulting policy and supplementary information, including experiments and demos, can be found on the website.

This paper explores the problem of collision-free motion generation for manipulators by formulating it as a global motion optimization problem. We develop a parallel optimization technique to solve this problem and demonstrate its effectiveness on massively parallel GPUs. We show that combining simple optimization techniques with many parallel seeds leads to solving difficult motion generation problems within 53ms on average, 62x faster than SOTA trajectory optimization methods. We achieve SOTA performance by combining L-BFGS step direction estimation with a novel parallel noisy line search scheme and a particle-based optimization solver. To further aid trajectory optimization, we develop a parallel geometric planner that is atleast 28x faster than SOTA RRTConnect implementations. We also introduce a collision-free IK solver that can solve over 9000 queries/s. We are releasing our GPU accelerated library CuRobo that contains core components for robot motion generation. Additional details are available at sites.google.com/nvidia.com/curobo.

Motion generation seeks to produce safe and feasible robot motion from start to goal. Various tools at different levels of granularity have been developed. On one extreme, sampling-based motion planners focus on completeness - a solution, if it exists, would eventually be found. However, produced paths are often of low quality, and contain superfluous motion. On the other, reactive methods optimise the immediate cost to obtain the next controls, producing smooth and legible motion that can quickly adapt to perturbations, uncertainties, and changes in the environment. However, reactive methods are highly local, and often produce motion that become trapped in non-convex regions of the environment. This paper contributes, Geometric Fabric Command Sequences, a method that lies in the middle ground. It can produce globally optimal motion that is smooth and intuitive, while being also reactive. We model motion via a reactive Geometric Fabric policy that ingests a sequence of attractor states, or commands, and then apply global optimisation over the space of commands. We postulate that solutions for different problems and scenes are highly transferable when conditioned on environmental features. Therefore, an implicit generative model is trained on solutions from optimisation and environment features in a self-supervised manner. That is, faced with multiple motion generation problems, the learning and optimisation are contained within the same loop: the optimisation generates labels for learning, while the learning improves the optimisation for the next problem, which in turn provides higher quality labels. We empirically validate our method in both simulation and on a real-world 6-DOF JACO arm.

We introduce DexYCB, a new dataset for capturing hand grasping of objects. We first compare DexYCB with a related one through cross-dataset evaluation. We then present a thorough benchmark of state-of-the-art approaches on three relevant tasks: 2D object and keypoint detection, 6D object pose estimation, and 3D hand pose estimation. Finally, we evaluate a new robotics-relevant task: generating safe robot grasps in human-to-robot object handover.1

Robotics research has found numerous important applications of Riemannian geometry. Despite that, the concept remain challenging to many roboticists because the background material is complex and strikingly foreign. Beyond Riemannian geometry, there are many natural generalizations in the mathematical literature—areas such as Finsler geometry and spray geometry—but those generalizations are largely inaccessible, and as a result there remain few applications within robotics. This paper presents a re-derivation of spray and Finsler geometries, critical for the development of our recent work on geometric fabrics, which builds the ideas from familiar concepts in advanced calculus and the calculus of variations. We focus on the pragmatic and calculable results, avoiding the use of tensor notation to appeal to a broader audience and emphasizing geometric path consistency over ideas around connections and curvature. We hope that they will contribute to an increased understanding of generalized nonlinear, and even classical Riemannian, geometry within the robotics community and inspire future research into new applications.

Teleoperation offers the possibility of imparting robotic systems with sophisticated reasoning skills, intuition, and creativity to perform tasks. However, teleoperation solutions for high degree-of-actuation (DoA), multi-fingered robots are generally cost-prohibitive, while low-cost offerings usually offer reduced degrees of control. Herein, a low-cost, depth-based teleoperation system, DexPilot, was developed that allows for complete control over the full 23 DoA robotic system by merely observing the bare human hand. DexPilot enabled operators to solve a variety of complex manipulation tasks that go beyond simple pick-and-place operations and performance was measured through speed and reliability metrics. DexPilot cost-effectively enables the production of high dimensional, multi-modality, state-action data that can be leveraged in the future to learn sensorimotor policies for challenging manipulation tasks. The videos of the experiments can be found at https://sites.google.com/view/dex-pilot.

Tracking the pose of an object while it is being held and manipulated by a robot hand is difficult for vision-based methods due to significant occlusions. Prior works have explored using contact feedback and particle filters to localize in-hand objects. However, they have mostly focused on the static grasp setting and not when the object is in motion, as doing so requires modeling of complex contact dynamics. In this work, we propose using GPU-accelerated parallel robot simulations and derivative-free, sample-based optimizers to track in-hand object poses with contact feedback during manipulation. We use physics simulation as the forward model for robot-object interactions, and the algorithm jointly optimizes for the state and the parameters of the simulations, so they better match with those of the real world. Our method runs in real-time (30Hz) on a single GPU, and it achieves an average point cloud distance error of 6mm in simulation experiments and 13mm in the real-world ones.

The existence of tactile afferents sensitive to slip-related mechanical transients in the human hand augments the robustness of grasping through secondary force modulation protocols. Despite this knowledge and the fact that tactile-based slip detection has been researched for decades, robust slip detection is still not an out-of-the-box capability for any commercially available tactile sensor. This research seeks to bridge this gap with a comprehensive study addressing several aspects of slip detection. In particular, key developments include a systematic data collection process yielding millions of sensory data points, a spectral analysis of sensory responses providing insight into sensor behavior, and the application of Long Short-Term Memory (LSTM) neural networks to produce robust slip detectors from three commercially available sensors capable of tactile sensing. The sensing mechanics behind these sensors are all fundamentally different and leverage principles in electro-mechanical resistance, optics, and hydro-acoustics. Critically, slip detection performance of the tactile technologies is quantified through a measurement methodology that unveils the effects of data window size, sampling rate, material type, slip speed, and sensor manufacturing variability. Results indicate that the investigated commercial tactile sensors are inherently capable of high-quality slip detection.

Automated grasping has a long history of research that is increasing due to interest from industry. One grand challenge for robotics is Universal Picking: the ability to robustly grasp a broad variety of objects in diverse environments for applications from warehouses to assembly lines to homes. Although many researchers now openly share code and data, it is challenging to compare and/or reproduce experimental results to identify which aspects of which approaches work best due to variations in assumptions and experimental protocols, e.g., sensors, lighting, robot arms, grippers, and objects.

The ability to calculate rigid-body transformations between arbitrary coordinate systems (i.e., registration) is an invaluable tool in robotics. This effort builds upon previous work by investigating strategies for improving the registration accuracy between a robotic arm and an extrinsic coordinate system with relatively inexpensive parts and minimal labor. The framework previously presented is expanded with a new test methodology to characterize the effects of strategies that improve registration performance. In addition, statistical analyses of physical trials reveal that leveraging more data and applying machine learning are two major components for significantly reducing registration error. One-shot peg-in-hole tests are conducted to show the application-level performance gains obtained by improving registration accuracy. Trends suggest that the maximum translation positioning error (postregistration) is a good, albeit not perfect, indicator for peg insertion performance. NOTE TO PRACTITIONERS: In a dynamic robotic workcell environment where robots may be frequently relocated or may need to collaborate with other robots, it is simpler and more robust to program robots in an external or unifying reference frame. The process of robot registration involves finding the location of a robot with respect to another reference frame. For instance, if parts are in known locations on a table and a robot can locate itself with respect to the table (an external reference frame), then the robot will also know the location of the parts. Furthermore, if two or more robots can locate themselves with respect to the table, then each robot will not only know the location of the parts, but also the location of every other robot. This knowledge facilitates the coordination of robot motions and robot collaboration and eases the integration of additional robots into the workcell. Since robot registration is critically necessary and occurs frequently, its process needs to be inexpensive, fast, and accurate. This paper details the requirements for a relatively inexpensive and fast robot registration experience, along with detailing strategies that incur significant improvements to registered robot positioning accuracy with minimal overhead. A quantitative verification process is presented to evaluate the performance impacts of these strategies. Peg-in-hole experiments are conducted to validate the notion that more accurate robot registration translates to more reliable task-level performance.

Force-based manipulation control strategies are evolving as a primary mechanism in robotics for performing the fine manipulation tasks typical within manufacturing assembly. The ability to systematically compare robotic system performance and quantify true advancement in fine manipulation is of utmost importance. Accordingly, the objectives of this paper are threefold: 1) creation of a peg-in-hole test method with associated performance metrics and a systematic data analysis strategy for performance benchmarking, 2) first demonstration of a recently developed manipulation controller piloting a robotic hand and its paired task-level logic for completing the peg-in-hole test, and 3) exemplifying the performance benchmarking technique by comparing two approaches for robotic insertions-the previously mentioned compliant hand, stiff arm system, and a stiff gripper, compliant arm system. Analyses reveal that the unconventional hand system can perform at and sometimes above the level of the gripper system in the developed peg-in-hole scenario. Moreover, the hand's active control of the peg's full Cartesian pose reduces positional error sensitivity and minimizes exerted insertion forces, highlighting the strategy's potential for fine manipulation tasks. Results indicate that robotic arms equipped with highly articulated and sensorized robotic hands can provide a truly realizable solution path for performing peg-in-hole tasks.