Real Robot Challenge 2022: Learning Dexterous Manipulation from Offline Data in the Real World

Experimentation on real robots is demanding in terms of time and costs. For this reason, a large part of the reinforcement learning (RL) community uses simulators to develop and benchmark algorithms. However, insights gained in simulation do not necessarily translate to real robots, in particular for tasks involving complex interactions with the environment. The Real Robot Challenge 2022 therefore served as a bridge between the RL and robotics communities by allowing participants to experiment remotely with a real robot - as easily as in simulation. In the last years, offline reinforcement learning has matured into a promising paradigm for learning from pre-collected datasets, alleviating the reliance on expensive online interactions. We therefore asked the participants to learn two dexterous manipulation tasks involving pushing, grasping, and in-hand orientation from provided real-robot datasets. An extensive software documentation and an initial stage based on a simulation of the real set-up made the competition particularly accessible. By giving each team plenty of access budget to evaluate their offline-learned policies on a cluster of seven identical real TriFinger platforms, we organized an exciting competition for machine learners and roboticists alike. In this work we state the rules of the competition, present the methods used by the winning teams and compare their results with a benchmark of state-of-the-art offline RL algorithms on the challenge datasets.

Benchmarking Offline Reinforcement Learning on Real-Robot Hardware

Learning policies from previously recorded data is a promising direction for real-world robotics tasks, as online learning is often infeasible. Dexterous manipulation in particular remains an open problem in its general form. The combination of offline reinforcement learning with large diverse datasets, however, has the potential to lead to a breakthrough in this challenging domain analogously to the rapid progress made in supervised learning in recent years. To coordinate the efforts of the research community toward tackling this problem, we propose a benchmark including: i) a large collection of data for offline learning from a dexterous manipulation platform on two tasks, obtained with capable RL agents trained in simulation; ii) the option to execute learned policies on a real-world robotic system and a simulation for efficient debugging. We evaluate prominent open-sourced offline reinforcement learning algorithms on the datasets and provide a reproducible experimental setup for offline reinforcement learning on real systems.

Synchronizing Machine Learning Algorithms, Realtime Robotic Control and Simulated Environment with o80

Robotic applications require the integration of various modalities, encompassing perception, control of real robots and possibly the control of simulated environments. While the state-of-the-art robotic software solutions such as ROS 2 provide most of the required features, flexible synchronization between algorithms, data streams and control loops can be tedious. o80 is a versatile C++ framework for robotics which provides a shared memory model and a command framework for real-time critical systems. It enables expert users to set up complex robotic systems and generate Python bindings for scientists. o80's unique feature is its flexible synchronization between processes, including the traditional blocking commands and the novel ``bursting mode'', which allows user code to control the execution of the lower process control loop. This makes it particularly useful for setups that mix real and simulated environments.

Dexterous Robotic Manipulation using Deep Reinforcement Learning and Knowledge Transfer for Complex Sparse Reward-based Tasks

This paper describes a deep reinforcement learning (DRL) approach that won Phase 1 of the Real Robot Challenge (RRC) 2021, and then extends this method to a more difficult manipulation task. The RRC consisted of using a TriFinger robot to manipulate a cube along a specified positional trajectory, but with no requirement for the cube to have any specific orientation. We used a relatively simple reward function, a combination of goal-based sparse reward and distance reward, in conjunction with Hindsight Experience Replay (HER) to guide the learning of the DRL agent (Deep Deterministic Policy Gradient (DDPG)). Our approach allowed our agents to acquire dexterous robotic manipulation strategies in simulation. These strategies were then applied to the real robot and outperformed all other competition submissions, including those using more traditional robotic control techniques, in the final evaluation stage of the RRC. Here we extend this method, by modifying the task of Phase 1 of the RRC to require the robot to maintain the cube in a particular orientation, while the cube is moved along the required positional trajectory. The requirement to also orient the cube makes the agent unable to learn the task through blind exploration due to increased problem complexity. To circumvent this issue, we make novel use of a Knowledge Transfer (KT) technique that allows the strategies learned by the agent in the original task (which was agnostic to cube orientation) to be transferred to this task (where orientation matters). KT allowed the agent to learn and perform the extended task in the simulator, which improved the average positional deviation from 0.134 m to 0.02 m, and average orientation deviation from 142{\deg} to 76{\deg} during evaluation. This KT concept shows good generalisation properties and could be applied to any actor-critic learning algorithm.

A Robot Cluster for Reproducible Research in Dexterous Manipulation

Dexterous manipulation remains an open problem in robotics. To coordinate efforts of the research community towards tackling this problem, we propose a shared benchmark. We designed and built robotic platforms that are hosted at the MPI-IS and can be accessed remotely. Each platform consists of three robotic fingers that are capable of dexterous object manipulation. Users are able to control the platforms remotely by submitting code that is executed automatically, akin to a computational cluster. Using this setup, i) we host robotics competitions, where teams from anywhere in the world access our platforms to tackle challenging tasks, ii) we publish the datasets collected during these competitions (consisting of hundreds of robot hours), and iii) we give researchers access to these platforms for their own projects.

Transferring Dexterous Manipulation from GPU Simulation to a Remote Real-World TriFinger

We present a system for learning a challenging dexterous manipulation task involving moving a cube to an arbitrary 6-DoF pose with only 3-fingers trained with NVIDIA's IsaacGym simulator. We show empirical benefits, both in simulation and sim-to-real transfer, of using keypoints as opposed to position+quaternion representations for the object pose in 6-DoF for policy observations and in reward calculation to train a model-free reinforcement learning agent. By utilizing domain randomization strategies along with the keypoint representation of the pose of the manipulated object, we achieve a high success rate of 83% on a remote TriFinger system maintained by the organizers of the Real Robot Challenge. With the aim of assisting further research in learning in-hand manipulation, we make the codebase of our system, along with trained checkpoints that come with billions of steps of experience available, at https://s2r2-ig.github.io

Representation Learning for Out-Of-Distribution Generalization in Reinforcement Learning

Learning data representations that are useful for various downstream tasks is a cornerstone of artificial intelligence. While existing methods are typically evaluated on downstream tasks such as classification or generative image quality, we propose to assess representations through their usefulness in downstream control tasks, such as reaching or pushing objects. By training over 10,000 reinforcement learning policies, we extensively evaluate to what extent different representation properties affect out-of-distribution (OOD) generalization. Finally, we demonstrate zero-shot transfer of these policies from simulation to the real world, without any domain randomization or fine-tuning. This paper aims to establish the first systematic characterization of the usefulness of learned representations for real-world OOD downstream tasks.

Benchmarking Structured Policies and Policy Optimization for Real-World Dexterous Object Manipulation

Dexterous manipulation is a challenging and important problem in robotics. While data-driven methods are a promising approach, current benchmarks require simulation or extensive engineering support due to the sample inefficiency of popular methods. We present benchmarks for the TriFinger system, an open-source robotic platform for dexterous manipulation and the focus of the 2020 Real Robot Challenge. The benchmarked methods, which were successful in the challenge, can be generally described as structured policies, as they combine elements of classical robotics and modern policy optimization. This inclusion of inductive biases facilitates sample efficiency, interpretability, reliability and high performance. The key aspects of this benchmarking is validation of the baselines across both simulation and the real system, thorough ablation study over the core features of each solution, and a retrospective analysis of the challenge as a manipulation benchmark. The code and demo videos for this work can be found on our website (https://sites.google.com/view/benchmark-rrc).

TriFinger: An Open-Source Robot for Learning Dexterity

Dexterous object manipulation remains an open problem in robotics, despite the rapid progress in machine learning during the past decade. We argue that a hindrance is the high cost of experimentation on real systems, in terms of both time and money. We address this problem by proposing an open-source robotic platform which can safely operate without human supervision. The hardware is inexpensive (about \SI{5000}[\$]{}) yet highly dynamic, robust, and capable of complex interaction with external objects. The software operates at 1-kilohertz and performs safety checks to prevent the hardware from breaking. The easy-to-use front-end (in C++ and Python) is suitable for real-time control as well as deep reinforcement learning. In addition, the software framework is largely robot-agnostic and can hence be used independently of the hardware proposed herein. Finally, we illustrate the potential of the proposed platform through a number of experiments, including real-time optimal control, deep reinforcement learning from scratch, throwing, and writing.

An Open Torque-Controlled Modular Robot Architecture for Legged Locomotion Research

We present a new open-source torque-controlled legged robot system, with a low cost and low complexity actuator module at its core. It consists of a low-weight high torque brushless DC motor and a low gear ratio transmission suitable for impedance and force control. We also present a novel foot contact sensor suitable for legged locomotion with hard impacts. A 2.2 kg quadruped robot with a large range of motion is assembled from 8 identical actuator modules and 4 lower legs with foot contact sensors. To the best of our knowledge, it is the most lightest force-controlled quadruped robot. We leverage standard plastic 3D printing and off-the-shelf parts, resulting in light-weight and inexpensive robots, allowing for rapid distribution and duplication within the research community. In order to quantify the capabilities of our design, we systematically measure the achieved impedance at the foot in static and dynamic scenarios. We measured up to 10.8 dimensionless leg stiffness without active damping, which is comparable to the leg stiffness of a running human. Finally, in order to demonstrate the capabilities of our quadruped robot, we propose a novel controller which combines feedforward contact forces computed from a kino-dynamic optimizer with impedance control of the robot center of mass and base orientation. The controller is capable of regulating complex motions which are robust to environmental uncertainty.