In this work we show how to use the Operational Space Control framework (OSC) under joint and cartesian constraints for reinforcement learning in cartesian space. Our method is therefore able to learn fast and with adjustable degrees of freedom, while we are able to transfer policies without additional dynamics randomizations on a KUKA LBR iiwa peg in-hole task. Before learning in simulation starts, we perform a system identification for aligning the simulation environment as far as possible with the dynamics of a real robot. Adding constraints to the OSC controller allows us to learn in a safe way on the real robot or to learn a flexible, goal conditioned policy that can be easily transferred from simulation to the real robot.
How to explore corner cases as efficiently and thoroughly as possible has long been one of the top concerns in the context of deep reinforcement learning (DeepRL) autonomous driving. Training with simulated data is less costly and dangerous than utilizing real-world data, but the inconsistency of parameter distribution and the incorrect system modeling in simulators always lead to an inevitable Sim2real gap, which probably accounts for the underperformance in novel, anomalous and risky cases that simulators can hardly generate. Domain Randomization(DR) is a methodology that can bridge this gap…
Producing agents that can generalize to a wide range of visually different environments is a significant challenge in reinforcement learning. One method for overcoming this issue is visual domain randomization, whereby at the start of each training episode some visual aspects of the environment are randomized so that the agent is exposed to many possible variations. However, domain randomization is highly inefficient and may lead to policies with high variance across domains. Instead, we propose a regularization method whereby the agent is only trained on one variation of the environment, and …
University of Cambridge（剑桥大学）；Helmut-Schmidt-University Hamburg（汉堡赫尔穆特·施密特大学）
Reinforcement learning frameworks have introduced abstractions to implement and execute algorithms at scale. They assume standardized simulator interfaces but are not concerned with identifying suitable task representations. We present Wield, a first-of-its kind system to facilitate task design for practical reinforcement learning. Through software primitives, Wield enables practitioners to decouple system-interface and deployment-specific configuration from state and action design. To guide experimentation, Wield further introduces a novel task design protocol and classification scheme centre…
Politecnico di Torino（都灵理工大学）；Aalto University（阿尔托大学）
In recent years, domain randomization over dynamics parameters has gained a lot of traction as a method for sim-to-real transfer of reinforcement learning policies in robotic manipulation; however, finding optimal randomization distributions can be difficult. In this paper, we introduce DROPO, a novel method for estimating domain randomization distributions for safe sim-to-real transfer. Unlike prior work, DROPO only requires a limited, precollected offline dataset of trajectories, and explicitly models parameter uncertainty to match real data using a likelihood-based approach. We demonstrate …
Recently, reinforcement learning (RL) algorithms have demonstrated remarkable success in learning complicated behaviors from minimally processed input. However, most of this success is limited to simulation. While there are promising successes in applying RL algorithms directly on real systems, their performance on more complex systems remains bottle-necked by the relative data inefficiency of RL algorithms. Domain randomization is a promising direction of research that has demonstrated impressive results using RL algorithms to control real robots. At a high level, domain randomization works b…
Visual domain randomization in simulated environments is a widely used method to transfer policies trained in simulation to real robots. However, domain randomization and augmentation hamper the training of a policy. As reinforcement learning struggles with a noisy training signal, this additional nuisance can drastically impede training. For difficult tasks it can even result in complete failure to learn. To overcome this problem we propose to pre-train a perception encoder that already provides an embedding invariant to the randomization. We demonstrate that this yields consistently improved…
Command and Control Engineering College, Army Engineering University of PLA（中国人民解放军陆军工程大学指挥控制工程学院）；troop of PLA（中国人民解放军31436部队）；Academy of Military Science（军事科学院）
The existing network attack and defense method can be regarded as game, but most of the game only involves network domain, not multiple domain cyberspace. To address this challenge, this paper proposed a multiple domain cyberspace attack and defense game model based on reinforcement learning. We define the multiple domain cyberspace include physical domain, network domain and digital domain. By establishing two agents, representing the attacker and the defender respectively, defender will select the multiple domain actions in the multiple domain cyberspace to obtain defender's optimal reward b…
KAIST（韩国科学技术院）；University of Michigan（密歇根大学）；Google Brain（谷歌大脑）
Deep reinforcement learning (RL) agents often fail to generalize to unseen environments (yet semantically similar to trained agents), particularly when they are trained on high-dimensional state spaces, such as images. In this paper, we propose a simple technique to improve a generalization ability of deep RL agents by introducing a randomized (convolutional) neural network that randomly perturbs input observations. It enables trained agents to adapt to new domains by learning robust features invariant across varied and randomized environments. Furthermore, we consider an inference method base…
NVIDIA（英伟达）；University of Southern California（南加利福尼亚大学）；University of Copenhagen（哥本哈根大学）；University of Washington（华盛顿大学）
We consider the problem of transferring policies to the real world by training on a distribution of simulated scenarios. Rather than manually tuning the randomization of simulations, we adapt the simulation parameter distribution using a few real world roll-outs interleaved with policy training. In doing so, we are able to change the distribution of simulations to improve the policy transfer by matching the policy behavior in simulation and the real world. We show that policies trained with our method are able to reliably transfer to different robots in two real world tasks: swing-peg-in-hole …
Reinforcement learning encounters many challenges when applied directly in the real world. Sim-to-real transfer is widely used to transfer the knowledge learned from simulation to the real world. Domain randomization -- one of the most popular algorithms for sim-to-real transfer -- has been demonstrated to be effective in various tasks in robotics and autonomous driving. Despite its empirical successes, theoretical understanding on why this simple algorithm works is limited. In this paper, we propose a theoretical framework for sim-to-real transfers, in which the simulator is modeled as a set …
Model-based Reinforcement Learning (MBRL) has been widely adapted due to its sample efficiency. However, existing worst-case regret analysis typically requires optimistic planning, which is not realistic in general. In contrast, motivated by the theory, empirical study utilizes ensemble of models, which achieve state-of-the-art performance on various testing environments. Such deviation between theory and empirical study leads us to question whether randomized model ensemble guarantee optimism, and hence the optimal worst-case regret? This paper partially answers such question from the perspec…
Transferring reinforcement learning policies trained in physics simulation to the real hardware remains a challenge, known as the "sim-to-real" gap. Domain randomization is a simple yet effective technique to address dynamics discrepancies across source and target domains, but its success generally depends on heuristics and trial-and-error. In this work we investigate the impact of randomized parameter selection on policy transferability across different types of domain discrepancies. Contrary to common practice in which kinematic parameters are carefully measured while dynamic parameters are …
We propose DeepSim, a reinforcement learning environment build toolkit for ROS and Gazebo. It allows machine learning or reinforcement learning researchers to access the robotics domain and create complex and challenging custom tasks in ROS and Gazebo simulation environments. This toolkit provides building blocks of advanced features such as collision detection, behaviour control, domain randomization, spawner, and many more. DeepSim is designed to reduce the boundary between robotics and machine learning communities by providing Python interface. In this paper, we discuss the components and d…
Deep reinforcement learning (DRL) policies are vulnerable to unauthorized replication attacks, where an adversary exploits imitation learning to reproduce target policies from observed behavior. In this paper, we propose Constrained Randomization of Policy (CRoP) as a mitigation technique against such attacks. CRoP induces the execution of sub-optimal actions at random under performance loss constraints. We present a parametric analysis of CRoP, address the optimality of CRoP, and establish theoretical bounds on the adversarial budget and the expectation of loss. Furthermore, we report the exp…
Stanford University（斯坦福大学）；NVIDIA（英伟达）；University of Sydney（悉尼大学）
BayesSim is a statistical technique for domain randomization in reinforcement learning based on likelihood-free inference of simulation parameters. This paper outlines BayesSimIG: a library that provides an implementation of BayesSim integrated with the recently released NVIDIA IsaacGym. This combination allows large-scale parameter inference with end-to-end GPU acceleration. Both inference and simulation get GPU speedup, with support for running more than 10K parallel simulation environments for complex robotics tasks that can have more than 100 simulation parameters to estimate. BayesSimIG p…
Mind the Gap: Towards Generalizable Autonomous Penetration Testing via Domain Randomization and Meta-Reinforcement Learning
College of Electronic Engineering, National University of Defense Technology（国防科技大学电子工程学院）；Beijing National Research Center for Information Science and Technology, Tsinghua University（清华大学北京信息科学与技术国家研究中心）；College of Computer Science and Technology, National University of Defense Technology（国防科技大学计算机学院）
Politecnico di Torino（都灵理工大学）；Aalto University（阿尔托大学）
Physics simulators have shown great promise for conveniently learning reinforcement learning policies in safe, unconstrained environments. However, transferring the acquired knowledge to the real world can be challenging due to the reality gap. To this end, several methods have been recently proposed to automatically tune simulator parameters with posterior distributions given real data, for use with domain randomization at training time. These approaches have been shown to work for various robotic tasks under different settings and assumptions. Nevertheless, existing literature lacks a thorou…
Mila（米拉）；Université de Montréal（蒙特利尔大学）；Element AI（元素人工智能公司）；CIFAR AI Chair（加拿大高级研究院人工智能讲席）
Goal-directed Reinforcement Learning (RL) traditionally considers an agent interacting with an environment, prescribing a real-valued reward to an agent proportional to the completion of some goal. Goal-directed RL has seen large gains in sample efficiency, due to the ease of reusing or generating new experience by proposing goals. One approach,self-play, allows an agent to "play" against itself by alternatively setting and accomplishing goals, creating a learned curriculum through which an agent can learn to accomplish progressively more difficult goals. However, self-play has been limited to…
Technical University of Darmstadt（达姆施塔特工业大学）；Honda Research Institute Europe（本田欧洲研究所）
Deep reinforcement learning is an effective tool to learn robot control policies from scratch. However, these methods are notorious for the enormous amount of required training data which is prohibitively expensive to collect on real robots. A highly popular alternative is to learn from simulations, allowing to generate the data much faster, safer, and cheaper. Since all simulators are mere models of reality, there are inevitable differences between the simulated and the real data, often referenced as the 'reality gap'. To bridge this gap, many approaches learn one policy from a distribution o…
Montreal Institute of Learning Algorithms (MILA)（蒙特利尔学习算法研究所（MILA））；McGill University（麦吉尔大学）；Facebook AI Research（Facebook人工智能研究院）
The goal of this work is to address the recent success of domain randomization and data augmentation for the sim2real setting. We explain this success through the lens of causal inference, positioning domain randomization and data augmentation as interventions on the environment which encourage invariance to irrelevant features. Such interventions include visual perturbations that have no effect on reward and dynamics. This encourages the learning algorithm to be robust to these types of variations and learn to attend to the true causal mechanisms for solving the task. This connection leads to…
Training deep reinforcement learning (DRL) locomotion policies often require massive amounts of data to converge to the desired behaviour. In this regard, simulators provide a cheap and abundant source. For successful sim-to-real transfer, exhaustively engineered approaches such as system identification, dynamics randomization, and domain adaptation are generally employed. As an alternative, we investigate a simple strategy of random force injection (RFI) to perturb system dynamics during training. We show that the application of random forces enables us to emulate dynamics randomization. This…
Safe Continual Domain Adaptation after Sim2Real Transfer of Reinforcement Learning Policies in Robotics
Technical University of Munich（慕尼黑工业大学）；University of California, Berkeley（加州大学伯克利分校）；Technische Universität Berlin（柏林工业大学）；L3S Research Center, Leibniz Universität Hannover（莱布尼茨汉诺威大学L3S研究中心）
University of British Columbia（不列颠哥伦比亚大学）；NVIDIA（英伟达）；University of Toronto（多伦多大学）；Vector Institute（向量研究院）
Understanding the gap between simulation and reality is critical for reinforcement learning with legged robots, which are largely trained in simulation. However, recent work has resulted in sometimes conflicting conclusions with regard to which factors are important for success, including the role of dynamics randomization. In this paper, we aim to provide clarity and understanding on the role of dynamics randomization in learning robust locomotion policies for the Laikago quadruped robot. Surprisingly, in contrast to prior work with the same robot model, we find that direct sim-to-real transf…
Autonomous Driving requires high levels of coordination and collaboration between agents. Achieving effective coordination in multi-agent systems is a difficult task that remains largely unresolved. Multi-Agent Reinforcement Learning has arisen as a powerful method to accomplish this task because it considers the interaction between agents and also allows for decentralized training -- which makes it highly scalable. However, transferring policies from simulation to the real world is a big challenge, even for single-agent applications. Multi-agent systems add additional complexities to the Sim-…
Université de Montréal（蒙特利尔大学）；Mila（米拉）；Polytechnique Montréal（蒙特利尔综合理工学院）；CIFAR AI Chair（加拿大高级研究院人工智能主席）
Gradient-based meta-learners such as Model-Agnostic Meta-Learning (MAML) have shown strong few-shot performance in supervised and reinforcement learning settings. However, specifically in the case of meta-reinforcement learning (meta-RL), we can show that gradient-based meta-learners are sensitive to task distributions. With the wrong curriculum, agents suffer the effects of meta-overfitting, shallow adaptation, and adaptation instability. In this work, we begin by highlighting intriguing failure cases of gradient-based meta-RL and show that task distributions can wildly affect algorithmic out…
National Chiao-Tung University（国立交通大学）；Pervasive Artificial Intelligence Research (PAIR) Labs（普及人工智能研究实验室）；Academia Sinica（中央研究院）；University of Alberta（阿尔伯塔大学）
Sim-to-real, a term that describes where a model is trained in a simulator then transferred to the real world, is a technique that enables faster deep reinforcement learning (DRL) training. However, differences between the simulator and the real world often cause the model to perform poorly in the real world. Domain randomization is a way to bridge the sim-to-real gap by exposing the model to a wide range of scenarios so that it can generalize to real-world situations. However, following domain randomization to train an autonomous car racing model with DRL can lead to undesirable outcomes. Nam…
Reinforcement learning has shown a wide usage in robotics tasks, such as insertion and grasping. However, without a practical sim2real strategy, the policy trained in simulation could fail on the real task. There are also wide researches in the sim2real strategies, but most of those methods rely on heavy image rendering, domain randomization training, or tuning. In this work, we solve the insertion task using a pure visual reinforcement learning solution with minimum infrastructure requirement. We also propose a novel sim2real strategy, Real2Sim, which provides a novel and easier solution in p…
School of Computation Information and Technology, Technical University of Munich（技术大学慕尼黑计算信息与技术学院）；Department of Computer Science, University of Innsbruck（因斯布鲁克大学计算机科学系）；L3S Research Center, Leibniz Universität Hannover（汉诺威莱布尼茨大学L3S研究中心）；Digital Science Center (DiSC), University of Innsbruck（因斯布鲁克大学数字科学中心）
UC Berkeley（加州大学伯克利分校）；Institute of Transportation Studies（交通研究所）
Reinforcement Learning (RL) is an effective tool for controller design but can struggle with issues of robustness, failing catastrophically when the underlying system dynamics are perturbed. The Robust RL formulation tackles this by adding worst-case adversarial noise to the dynamics and constructing the noise distribution as the solution to a zero-sum minimax game. However, existing work on learning solutions to the Robust RL formulation has primarily focused on training a single RL agent against a single adversary. In this work, we demonstrate that using a single adversary does not consisten…
Politecnico di Torino（都灵理工大学）；Technische Universität Darmstadt（达姆施塔特工业大学）；Hessian Center for Artificial Intelligence (Hessian.AI)（黑森人工智能中心）；German Research Center for AI (DFKI)（德国人工智能研究中心）；University of Würzburg（维尔茨堡大学）
We propose a method to predict the sim-to-real transfer performance of RL policies. Our transfer metric simplifies the selection of training setups (such as algorithm, hyperparameters, randomizations) and policies in simulation, without the need for extensive and time-consuming real-world rollouts. A probabilistic dynamics model is trained alongside the policy and evaluated on a fixed set of real-world trajectories to obtain the transfer metric. Experiments show that the transfer metric is highly correlated with policy performance in both simulated and real-world robotic environments for compl…
Accelerating learning processes for complex tasks by leveraging previously learned tasks has been one of the most challenging problems in reinforcement learning, especially when the similarity between source and target tasks is low. This work proposes REPresentation And INstance Transfer (REPAINT) algorithm for knowledge transfer in deep reinforcement learning. REPAINT not only transfers the representation of a pre-trained teacher policy in the on-policy learning, but also uses an advantage-based experience selection approach to transfer useful samples collected following the teacher policy in…
Mean Field Games (MFG) are the class of games with a very large number of agents and the standard equilibrium concept is a Mean Field Equilibrium (MFE). Algorithms for learning MFE in dynamic MFGs are unknown in general. Our focus is on an important subclass that possess a monotonicity property called Strategic Complementarities (MFG-SC). We introduce a natural refinement to the equilibrium concept that we call Trembling-Hand-Perfect MFE (T-MFE), which allows agents to employ a measure of randomization while accounting for the impact of such randomization on their payoffs. We propose a simple …
Shenyang Institute of Automation, Chinese Academy of Sciences（中国科学院沈阳自动化研究所）；Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences（中国科学院机器人与智能制造创新研究院）；University of Chinese Academy of Sciences（中国科学院大学）；Heriot-Watt University（赫瑞-瓦特大学）
Agile control of mobile manipulator is challenging because of the high complexity coupled by the robotic system and the unstructured working environment. Tracking and grasping a dynamic object with a random trajectory is even harder. In this paper, a multi-task reinforcement learning-based mobile manipulation control framework is proposed to achieve general dynamic object tracking and grasping. Several basic types of dynamic trajectories are chosen as the task training set. To improve the policy generalization in practice, random noise and dynamics randomization are introduced during the train…
Panasonic Beta（松下Beta）；Totemic Inc.（Totemic公司）；University of California Berkeley（加州大学伯克利分校）
In order to practically implement the door opening task, a policy ought to be robust to a wide distribution of door types and environment settings. Reinforcement Learning (RL) with Domain Randomization (DR) is a promising technique to enforce policy generalization, however, there are only a few accessible training environments that are inherently designed to train agents in domain randomized environments. We introduce DoorGym, an open-source door opening simulation framework designed to utilize domain randomization to train a stable policy. We intend for our environment to lie at the intersect…
Eötvös Loránd Research Network（厄特沃什·罗兰研究网络）；Institute for Computer Science and Control（计算机科学与控制研究所）；Eötvös Loránd University（厄特沃什·罗兰大学）；Budapest University of Technology and Economics（布达佩斯技术与经济大学）；Pavol Jozef Šafárik University（帕沃尔·约瑟夫·沙法里克大学）
Robots working in unstructured environments must be capable of sensing and interpreting their surroundings. One of the main obstacles of deep-learning-based models in the field of robotics is the lack of domain-specific labeled data for different industrial applications. In this article, we propose a sim2real transfer learning method based on domain randomization for object detection with which labeled synthetic datasets of arbitrary size and object types can be automatically generated. Subsequently, a state-of-the-art convolutional neural network, YOLOv4, is trained to detect the different ty…
University of California, San Diego（加利福尼亚大学圣迭戈分校）；Tsinghua University（清华大学）
Developing robust vision-guided controllers for quadrupedal robots in complex environments, with various obstacles, dynamical surroundings and uneven terrains, is very challenging. While Reinforcement Learning (RL) provides a promising paradigm for agile locomotion skills with vision inputs in simulation, it is still very challenging to deploy the RL policy in the real world. Our key insight is that aside from the discrepancy in the domain gap, in visual appearance between the simulation and the real world, the latency from the control pipeline is also a major cause of difficulty. In this pape…
