Scaling Population-Based Reinforcement Learning with GPU Accelerated Simulation

In recent years, deep reinforcement learning (RL) has shown its effectiveness in solving complex continuous control tasks like locomotion and dexterous manipulation. However, this comes at the cost of an enormous amount of experience required for training, exacerbated by the sensitivity of learning efficiency and the policy performance to hyperparameter selection, which often requires numerous trials of time-consuming experiments. This work introduces a Population-Based Reinforcement Learning (PBRL) approach that exploits a GPU-accelerated physics simulator to enhance the exploration capabilities of RL by concurrently training multiple policies in parallel. The PBRL framework is applied to three state-of-the-art RL algorithms -- PPO, SAC, and DDPG -- dynamically adjusting hyperparameters based on the performance of learning agents. The experiments are performed on four challenging tasks in Isaac Gym -- Anymal Terrain, Shadow Hand, Humanoid, Franka Nut Pick -- by analyzing the effect of population size and mutation mechanisms for hyperparameters. The results show that PBRL agents achieve superior performance, in terms of cumulative reward, compared to non-evolutionary baseline agents. The trained agents are finally deployed in the real world for a Franka Nut Pick task, demonstrating successful sim-to-real transfer. Code and videos of the learned policies are available on our project website.

MimicGen: A Data Generation System for Scalable Robot Learning using Human Demonstrations

Imitation learning from a large set of human demonstrations has proved to be an effective paradigm for building capable robot agents. However, the demonstrations can be extremely costly and time-consuming to collect. We introduce MimicGen, a system for automatically synthesizing large-scale, rich datasets from only a small number of human demonstrations by adapting them to new contexts. We use MimicGen to generate over 50K demonstrations across 18 tasks with diverse scene configurations, object instances, and robot arms from just ~200 human demonstrations. We show that robot agents can be effectively trained on this generated dataset by imitation learning to achieve strong performance in long-horizon and high-precision tasks, such as multi-part assembly and coffee preparation, across broad initial state distributions. We further demonstrate that the effectiveness and utility of MimicGen data compare favorably to collecting additional human demonstrations, making it a powerful and economical approach towards scaling up robot learning. Datasets, simulation environments, videos, and more at https://mimicgen.github.io .

IndustReal: Transferring Contact-Rich Assembly Tasks from Simulation to Reality

Robotic assembly is a longstanding challenge, requiring contact-rich interaction and high precision and accuracy. Many applications also require adaptivity to diverse parts, poses, and environments, as well as low cycle times. In other areas of robotics, simulation is a powerful tool to develop algorithms, generate datasets, and train agents. However, simulation has had a more limited impact on assembly. We present IndustReal, a set of algorithms, systems, and tools that solve assembly tasks in simulation with reinforcement learning (RL) and successfully achieve policy transfer to the real world. Specifically, we propose 1) simulation-aware policy updates, 2) signed-distance-field rewards, and 3) sampling-based curricula for robotic RL agents. We use these algorithms to enable robots to solve contact-rich pick, place, and insertion tasks in simulation. We then propose 4) a policy-level action integrator to minimize error at policy deployment time. We build and demonstrate a real-world robotic assembly system that uses the trained policies and action integrator to achieve repeatable performance in the real world. Finally, we present hardware and software tools that allow other researchers to reproduce our system and results. For videos and additional details, please see http://sites.google.com/nvidia.com/industreal .

DefGraspNets: Grasp Planning on 3D Fields with Graph Neural Nets

Robotic grasping of 3D deformable objects is critical for real-world applications such as food handling and robotic surgery. Unlike rigid and articulated objects, 3D deformable objects have infinite degrees of freedom. Fully defining their state requires 3D deformation and stress fields, which are exceptionally difficult to analytically compute or experimentally measure. Thus, evaluating grasp candidates for grasp planning typically requires accurate, but slow 3D finite element method (FEM) simulation. Sampling-based grasp planning is often impractical, as it requires evaluation of a large number of grasp candidates. Gradient-based grasp planning can be more efficient, but requires a differentiable model to synthesize optimal grasps from initial candidates. Differentiable FEM simulators may fill this role, but are typically no faster than standard FEM. In this work, we propose learning a predictive graph neural network (GNN), DefGraspNets, to act as our differentiable model. We train DefGraspNets to predict 3D stress and deformation fields based on FEM-based grasp simulations. DefGraspNets not only runs up to 1500 times faster than the FEM simulator, but also enables fast gradient-based grasp optimization over 3D stress and deformation metrics. We design DefGraspNets to align with real-world grasp planning practices and demonstrate generalization across multiple test sets, including real-world experiments.

DeXtreme: Transfer of Agile In-hand Manipulation from Simulation to Reality

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 at \url{https://dextreme.org/}

Factory: Fast Contact for Robotic Assembly

Robotic assembly is one of the oldest and most challenging applications of robotics. In other areas of robotics, such as perception and grasping, simulation has rapidly accelerated research progress, particularly when combined with modern deep learning. However, accurately, efficiently, and robustly simulating the range of contact-rich interactions in assembly remains a longstanding challenge. In this work, we present Factory, a set of physics simulation methods and robot learning tools for such applications. We achieve real-time or faster simulation of a wide range of contact-rich scenes, including simultaneous simulation of 1000 nut-and-bolt interactions. We provide $60$ carefully-designed part models, 3 robotic assembly environments, and 7 robot controllers for training and testing virtual robots. Finally, we train and evaluate proof-of-concept reinforcement learning policies for nut-and-bolt assembly. We aim for Factory to open the doors to using simulation for robotic assembly, as well as many other contact-rich applications in robotics. Please see https://sites.google.com/nvidia.com/factory for supplementary content, including videos.

Accelerated Policy Learning with Parallel Differentiable Simulation

Deep reinforcement learning can generate complex control policies, but requires large amounts of training data to work effectively. Recent work has attempted to address this issue by leveraging differentiable simulators. However, inherent problems such as local minima and exploding/vanishing numerical gradients prevent these methods from being generally applied to control tasks with complex contact-rich dynamics, such as humanoid locomotion in classical RL benchmarks. In this work we present a high-performance differentiable simulator and a new policy learning algorithm (SHAC) that can effectively leverage simulation gradients, even in the presence of non-smoothness. Our learning algorithm alleviates problems with local minima through a smooth critic function, avoids vanishing/exploding gradients through a truncated learning window, and allows many physical environments to be run in parallel. We evaluate our method on classical RL control tasks, and show substantial improvements in sample efficiency and wall-clock time over state-of-the-art RL and differentiable simulation-based algorithms. In addition, we demonstrate the scalability of our method by applying it to the challenging high-dimensional problem of muscle-actuated locomotion with a large action space, achieving a greater than 17x reduction in training time over the best-performing established RL algorithm.

DefGraspSim: Physics-based simulation of grasp outcomes for 3D deformable objects

Robotic grasping of 3D deformable objects (e.g., fruits/vegetables, internal organs, bottles/boxes) is critical for real-world applications such as food processing, robotic surgery, and household automation. However, developing grasp strategies for such objects is uniquely challenging. Unlike rigid objects, deformable objects have infinite degrees of freedom and require field quantities (e.g., deformation, stress) to fully define their state. As these quantities are not easily accessible in the real world, we propose studying interaction with deformable objects through physics-based simulation. As such, we simulate grasps on a wide range of 3D deformable objects using a GPU-based implementation of the corotational finite element method (FEM). To facilitate future research, we open-source our simulated dataset (34 objects, 1e5 Pa elasticity range, 6800 grasp evaluations, 1.1M grasp measurements), as well as a code repository that allows researchers to run our full FEM-based grasp evaluation pipeline on arbitrary 3D object models of their choice. Finally, we demonstrate good correspondence between grasp outcomes on simulated objects and their real counterparts.

DiSECt: A Differentiable Simulator for Parameter Inference and Control in Robotic Cutting

Robotic cutting of soft materials is critical for applications such as food processing, household automation, and surgical manipulation. As in other areas of robotics, simulators can facilitate controller verification, policy learning, and dataset generation. Moreover, differentiable simulators can enable gradient-based optimization, which is invaluable for calibrating simulation parameters and optimizing controllers. In this work, we present DiSECt: the first differentiable simulator for cutting soft materials. The simulator augments the finite element method (FEM) with a continuous contact model based on signed distance fields (SDF), as well as a continuous damage model that inserts springs on opposite sides of the cutting plane and allows them to weaken until zero stiffness, enabling crack formation. Through various experiments, we evaluate the performance of the simulator. We first show that the simulator can be calibrated to match resultant forces and deformation fields from a state-of-the-art commercial solver and real-world cutting datasets, with generality across cutting velocities and object instances. We then show that Bayesian inference can be performed efficiently by leveraging the differentiability of the simulator, estimating posteriors over hundreds of parameters in a fraction of the time of derivative-free methods. Next, we illustrate that control parameters in the simulation can be optimized to minimize cutting forces via lateral slicing motions. Finally, we conduct experiments on a real robot arm equipped with a slicing knife to infer simulation parameters from force measurements. By optimizing the slicing motion of the knife, we show on fruit cutting scenarios that the average knife force can be reduced by more than 40% compared to a vertical cutting motion. We publish code and additional materials on our project website at https://diff-cutting-sim.github.io.

Assistive Tele-op: Leveraging Transformers to Collect Robotic Task Demonstrations

Sharing autonomy between robots and human operators could facilitate data collection of robotic task demonstrations to continuously improve learned models. Yet, the means to communicate intent and reason about the future are disparate between humans and robots. We present Assistive Tele-op, a virtual reality (VR) system for collecting robot task demonstrations that displays an autonomous trajectory forecast to communicate the robot's intent. As the robot moves, the user can switch between autonomous and manual control when desired. This allows users to collect task demonstrations with both a high success rate and with greater ease than manual teleoperation systems. Our system is powered by transformers, which can provide a window of potential states and actions far into the future -- with almost no added computation time. A key insight is that human intent can be injected at any location within the transformer sequence if the user decides that the model-predicted actions are inappropriate. At every time step, the user can (1) do nothing and allow autonomous operation to continue while observing the robot's future plan sequence, or (2) take over and momentarily prescribe a different set of actions to nudge the model back on track. We host the videos and other supplementary material at https://sites.google.com/view/assistive-teleop.