We present a novel method for simultaneous learning of depth, egomotion, object motion, and camera intrinsics from monocular videos, using only consistency across neighboring video frames as supervision signal. Similarly to prior work, our method learns by applying differentiable warping to frames and comparing the result to adjacent ones, but it provides several improvements: We address occlusions geometrically and differentiably, directly using the depth maps as predicted during training. We introduce randomized layer normalization, a novel powerful regularizer, and we account for object motion relative to the scene. To the best of our knowledge, our work is the first to learn the camera intrinsic parameters, including lens distortion, from video in an unsupervised manner, thereby allowing us to extract accurate depth and motion from arbitrary videos of unknown origin at scale. We evaluate our results on the Cityscapes, KITTI and Eu-RoC datasets, establishing new state of the art on depth prediction and odometry, and demonstrate qualitatively that depth prediction can be learned from a collection of YouTube videos.

Abstract-We present differentiable particle filters (DPFs): a differentiable implementation of the particle filter algorithm with learnable motion and measurement models. Since DPFs are end-to-end differentiable, we can efficiently train their models by optimizing end-to-end state estimation performance, rather than proxy objectives such as model accuracy. DPFs encode the structure of recursive state estimation with prediction and measurement update that operate on a probability distribution over states. This structure represents an algorithmic prior that improves learning performance in state estimation problems while enabling explainability of the learned model. Our experiments on simulated and real data show substantial benefits from end-toend learning with algorithmic priors, e.g. reducing error rates by ∼80%. Our experiments also show that, unlike long short-term memory networks, DPFs learn localization in a policy-agnostic way and thus greatly improve generalization. Source code is available at https://github.com/tu-rbo/differentiable-particle-filters.

Robot learning is critically enabled by the availability of appropriate state representations. We propose a robotics-specific approach to learning such state representations. As robots accomplish tasks by interacting with the physical world, we can facilitate representation learning by considering the structure imposed by physics; this structure is reflected in the changes that occur in the world and in the way a robot can effect them. By exploiting this structure in learning, robots can obtain state representations consistent with the aspects of physics relevant to the learning task. We name this prior knowledge about the structure of interactions with the physical world robotic priors. We identify five robotic priors and explain how they can be used to learn pertinent state representations. We demonstrate the effectiveness of this approach in simulated and real robotic experiments with distracting moving objects. We show that our method extracts task-relevant state representations from high-dimensional observations, even in the presence of taskirrelevant distractions. We also show that the state representations learned by our method greatly improve generalization in reinforcement learning.

No abstract

We describe the winning entry to the Amazon Picking Challenge 2015. From the experience of building this system and competing, we derive several conclusions: (1) We suggest to characterize robotic system building along four key aspects, each of them spanning a spectrum of solutions - modularity vs. integration, generality vs. assumptions, computation vs. embodiment, and planning vs. feedback. (2) To understand which region of each spectrum most adequately addresses which robotic problem, we must explore the full spectrum of possible approaches. (3) For manipulation problems in unstructured environments, certain regions of each spectrum match the problem most adequately, and should be exploited further. This is supported by the fact that our solution deviated from the majority of the other challenge entries along each of the spectra. This is an abridged version of a conference publication.