Deep Reinforcement Learning (DRL) is emerging as a promising approach to generate adaptive behaviors for robotic platforms. However, a major drawback of using DRL is the data-hungry training regime that requires millions of trial and error attempts, which is impractical when running experiments on robotic systems. To address this issue, we propose a multi-subtask reinforcement learning method where complex tasks are decomposed manually into low-level subtasks by leveraging human domain knowledge. These subtasks can be parametrized as expert networks and learned via existing DRL methods. Trained subtasks can then be composed with a high-level choreographer. As a testbed, we use a pick and place robotic simulator to demonstrate our methodology, and show that our method outperforms an imitation learning-based method and reaches a high success rate compared to an endto-end learning approach. Moreover, we transfer the learned behavior in a different robotic environment that allows us to exploit sim-to-real transfer and demonstrate the trajectories in a real robotic system. Our training regime is carried out using a central processing unit (CPU)-based system, which demonstrates the data-efficient properties of our approach.

No abstract

Deep reinforcement learning (DRL) has become a dominant deep-learning paradigm for various tasks in which complex policies are learned within reactive systems. In parallel, there has recently been significant research on verifying deep neural networks. However, to date, there has been little work demonstrating the use of modern verification tools on real, DRLcontrolled systems. In this case-study paper, we attempt to begin bridging this gap, and focus on the important task of mapless robotic navigation -a classic robotics problem, in which a robot, usually controlled by a DRL agent, needs to efficiently and safely navigate through an unknown arena towards a desired target. We demonstrate how modern verification engines can be used for effective model selection, i.e., the process of selecting the best available policy for the robot in question from a pool of candidate policies. Specifically, we use verification to detect and rule out policies that may demonstrate suboptimal behavior, such as collisions and infinite loops. We also apply verification to identify models with overly conservative behavior, thus allowing users to choose superior policies that are better at finding an optimal, shorter path to a target. To validate our work, we conducted extensive experiments on an actual robot, and confirmed that the suboptimal policies detected by our method were indeed flawed. We also compared our verification-driven approach to state-of-the-art gradient attacks, and our results demonstrate that gradient-based methods are inadequate in this setting. Our work is the first to demonstrate the use of DNN verification backends for recognizing suboptimal DRL policies in real-world robots, and for filtering out unwanted policies. We believe that the methods presented in this work can be applied to a large range of application domains that incorporate deep-learning-based agents.

No abstract

Deep reinforcement learning (DRL) has become a dominant deep-learning paradigm for tasks where complex policies are learned within reactive systems. Unfortunately, these policies are known to be susceptible to bugs. Despite significant progress in DNN verification, there has been little work demonstrating the use of modern verification tools on real-world, DRL-controlled systems. In this case study, we attempt to begin bridging this gap, and focus on the important task of mapless robotic navigation — a classic robotics problem, in which a robot, usually controlled by a DRL agent, needs to efficiently and safely navigate through an unknown arena towards a target. We demonstrate how modern verification engines can be used for effective model selection, i.e., selecting the best available policy for the robot in question from a pool of candidate policies. Specifically, we use verification to detect and rule out policies that may demonstrate suboptimal behavior, such as collisions and infinite loops. We also apply verification to identify models with overly conservative behavior, thus allowing users to choose superior policies, which might be better at finding shorter paths to a target. To validate our work, we conducted extensive experiments on an actual robot, and confirmed that the suboptimal policies detected by our method were indeed flawed. We also demonstrate the superiority of our verification-driven approach over state-of-the-art, gradient attacks. Our work is the first to establish the usefulness of DNN verification in identifying and filtering out suboptimal DRL policies in real-world robots, and we believe that the methods presented here are applicable to a wide range of systems that incorporate deep-learning-based agents.