Autonomous quadrotor obstacle avoidance based on dueling double deep recurrent Q-learning with monocular vision
School of Aeronautic Science and Engineering, Beihang University（北京航空航天大学航空科学与工程学院）；Research Institute for Frontier Science, Beihang University（北京航空航天大学前沿科学研究院）；School of Electronic and Information Engineering, Beihang University（北京航空航天大学电子信息工程学院）
The rapid development of unmanned aerial vehicles (UAV) puts forward a higher requirement for autonomous obstacle avoidance. Due to the limited payload and power supply, small UAVs such as quadrotors usually carry simple sensors and computation units, which makes traditional methods more challenging to implement. In this paper, a novel framework is demonstrated to control a quadrotor flying through crowded environments autonomously with monocular vision. The framework adopts a two-stage architecture, consisting of a sensing module and a decision module. The sensing module is based on an unsupervised deep learning method. And the decision module uses dueling double deep recurrent Q-learning to eliminate the adverse effects of limited observation capacity of an on-board monocular camera. The framework enables the quadrotor to realize autonomous obstacle avoidance without any prior environment information or labeled datasets for training. The trained model shows a high success rate in the simulation and a good generalization ability for transformed scenarios.
University of Maryland Institute for Advanced Computer Studies（马里兰大学高级计算机研究所）
Although quadrotors, and aerial robots in general, are inherently active agents, their perceptual capabilities in literature so far have been mostly passive in nature. Researchers and practitioners today use traditional computer vision algorithms with the aim of building a representation of general applicability: a 3D reconstruction of the scene. Using this representation, planning tasks are constructed and accomplished to allow the quadrotor to demonstrate autonomous behavior. These methods are inefficient as they are not task driven and such methodologies are not utilized by flying insects a…
Ultra-low Power Deep Learning-based Monocular Relative Localization Onboard Nano-quadrotors
Dalle Molle Institute for Artificial Intelligence, USI and SUPSI（达勒·莫勒人工智能研究所（瑞士意大利语区大学与瑞士应用科学大学））；Integrated Systems Laboratory, ETH Zürich（苏黎世联邦理工学院集成系统实验室）
As the market for commercially available unmanned aerial vehicles (UAVs) booms, there is an increasing number of small, teleoperated or autonomous aircraft found in protected or sensitive airspace. Existing solutions for removal of these aircraft are either military-grade and too disruptive for domestic use, or compose of cumbersomely teleoperated counter-UAV vehicles that have proven ineffective in high-profile domestic cases. In this work, we examine the use of a quadrotor for autonomously targeting semi-stationary and moving aerial objects with little or no prior knowledge of the target's f…
The main contribution of this paper is a high frequency, low-complexity, on-board visual-inertial odometry system for quadrotor micro air vehicles. The system consists of an extended Kalman filter (EKF) based state estimation algorithm that fuses information from a low cost MEMS inertial measurement unit acquired at 200Hz and VGA resolution images from a monocular camera at 50Hz. The dynamic model describing the quadrotor motion is employed in the estimation algorithm as a third source of information. Visual information is incorporated into the EKF by enforcing the epipolar constraint on featu…
Hong Kong University of Science and Technology（香港科技大学）
We focus on a replanning scenario for quadrotors where considering time efficiency, non-static initial state and dynamical feasibility is of great significance. We propose a real-time B-spline based kinodynamic (RBK) search algorithm, which transforms a position-only shortest path search (such as A* and Dijkstra) into an efficient kinodynamic search, by exploring the properties of B-spline parameterization. The RBK search is greedy and produces a dynamically feasible time-parameterized trajectory efficiently, which facilitates non-static initial state of the quadrotor. To cope with the limitat…
Cameras provide a rich source of information while being passive, cheap and lightweight for small and medium Unmanned Aerial Vehicles (UAVs). In this work we present the first implementation of receding horizon control, which is widely used in ground vehicles, with monocular vision as the only sensing mode for autonomous UAV flight in dense clutter. We make it feasible on UAVs via a number of contributions: novel coupling of perception and control via relevant and diverse, multiple interpretations of the scene around the robot, leveraging recent advances in machine learning to showcase anytime…
ESIEA（ESIEA工程师学院）；ONERA - The French Aerospace Lab（法国航空航天实验室）
In this paper we address the problem of dot painting on a wall by a quadrotor Micro Air Vehicle (MAV), using on-board low cost sensors (monocular camera and IMU) for localization. A method is proposed to cope with uncertainties on the initial positioning of the MAV with respect to the wall and to deal with walls composed of multiple segments. This method is based on an online estimation algorithm that makes use of information of physical contacts detected by the drone during the flight to improve the positioning accuracy of the painted dots. Simulation results are presented to assess quantitat…
This paper presents our method for enabling a UAV quadrotor, equipped with a monocular camera, to autonomously avoid collisions with obstacles in unstructured and unknown indoor environments. When compared to obstacle avoidance in ground vehicular robots, UAV navigation brings in additional challenges because the UAV motion is no more constrained to a well-defined indoor ground or street environment. Horizontal structures in indoor and outdoor environments like decorative items, furnishings, ceiling fans, sign-boards, tree branches etc., also become relevant obstacles unlike those for ground v…
We present a real-time approach for image-based localization within large scenes that have been reconstructed offline using structure from motion (Sfm). From monocular video, our method continuously computes a precise 6-DOF camera pose, by efficiently tracking natural features and matching them to 3D points in the Sfm point cloud. Our main contribution lies in efficiently interleaving a fast keypoint tracker that uses inexpensive binary feature descriptors with a new approach for direct 2D-to-3D matching. The 2D-to-3D matching avoids the need for online extraction of scale-invariant features. …
