The potential benefits of model-free reinforcement learning to real robotics systems are limited by its uninformed exploration that leads to slow convergence, lack of data-efficiency, and unnecessary interactions with the environment. To address these drawbacks we propose a method that combines reinforcement and imitation learning by shaping the reward function with a state-and-action-dependent potential that is trained from demonstration data, using a generative model. We show that this accelerates policy learning by specifying high-value areas of the state and action space that are worth exploring first. Unlike the majority of existing methods that assume optimal demonstrations and incorporate the demonstration data as hard constraints on policy optimization, we instead incorporate demonstration data as advice in the form of a reward shaping potential trained as a generative model of states and actions. In particular, we examine both normalizing flows and Generative Adversarial Networks to represent these potentials. We show that, unlike many existing approaches that incorporate demonstrations as hard constraints, our approach is unbiased even in the case of suboptimal and noisy demonstrations. We present an extensive range of simulations, as well as experiments on the Franka Emika 7DOF arm, to demonstrate the practicality of our method.

To Perform reliably and consistently over sustained periods of time, large-scale automation critically relies on computer simulation. Simulation allows us and supervisory AI to effectively design, validate, and continuously improve complex processes, and helps practitioners to gain insight into the operation and justify future investments. While numerous successful applications of simulation in industry exist, such as circuit simulation, finite element methods, and computeraided design (CAD), state-of-the-art simulators fall short of accurately modeling physical phenomena, such as friction, impact, and deformation.

This paper considers the problem of learning behaviors in simulation without knowledge of the precise dynamical properties of the target robot platform(s). In this context, our learning goal is to mutually maximize task efficacy on each environment considered and generalization across the widest possible range of environmental conditions. The physical parameters of the simulator are modified by a component of our technique that learns the Domain Randomization (DR) that is appropriate at each learning epoch to maximally challenge the current behavior policy, without being overly challenging, which can hinder learning progress. This so-called sweet spot distribution is a selection of simulated domains with the following properties: 1) The trained policy should be successful in environments sampled from the domain randomization distribution; and 2) The DR distribution made as wide as possible, to increase variability in the environments. These properties aim to ensure the trajectories encountered in the target system are close to those observed during training, as existing methods in machine learning are better suited for interpolation than extrapolation. We show how adapting the DR distribution while training context-conditioned policies results in improvements on jump-start and asymptotic performance when transferring a learned policy to the target environment1.

We develop a multi-step motion prediction modeling method for dynamic systems over long horizons using deep learning. Building on previous work, we propose a novel hybrid network architecture, by combining deep recurrent neural networks with a quadrotor motion model created using classic system identification methods. The proposed model takes only the initial system state and motor speeds over the prediction horizon as inputs and returns robust state predictions for up to two seconds of motion at 100 Hz. We employ recurrent neural network state initialization during training, to exploit real-world dataset collected from quadrotor vehicle flights in an indoor flight arena. Our experiments demonstrate that the proposed hybrid network model consistently outperforms both black box and rigid body dynamics predictions over single and multi-step prediction scenarios, with an order of magnitude improvements in velocity estimates in particular.

We present AVOD, an Aggregate View Object Detection network for autonomous driving scenarios. The proposed neural network architecture uses LIDAR point clouds and RGB images to generate features that are shared by two subnetworks: a region proposal network (RPN) and a second stage detector network. The proposed RPN uses a novel architecture capable of performing multimodal feature fusion on high resolution feature maps to generate reliable 3D object proposals for multiple object classes in road scenes. Using these proposals, the second stage detection network performs accurate oriented 3D bounding box regression and category classification to predict the extents, orientation, and classification of objects in 3D space. Our proposed architecture is shown to produce state of the art results on the KITTI 3D object detection benchmark [1] while running in real time with a low memory footprint, making it a suitable candidate for deployment on autonomous vehicles. Code is available at: https://github.com/kujason/avod.