The days where swarms of unmanned aerial vehicles (UAVs) will occupy our skies are fast approaching due to the introduction of cost-efficient and reliable small aerial vehicles and the increasing demand for use of such vehicles in a plethora of civil applications. Governments and industry alike have been heavily investing in the development of UAVs. As such it is important to understand the characteristics of networks with UAVs to enable the incorporation of multiple, coordinated aerial vehicles into the air traffic in a reliable and safe manner. To this end, this survey reports the characteristics and requirements of UAV networks for envisioned civil applications over the period 2000–2015 from a communications and networking viewpoint. We survey and quantify quality-of-service requirements, network-relevant mission parameters, data requirements, and the minimum data to be transmitted over the network. Furthermore, we elaborate on general networking related requirements such as connectivity, adaptability, safety, privacy, security, and scalability. We also report experimental results from many projects and investigate the suitability of existing communication technologies for supporting reliable aerial networking.

Survey on Unmanned Aerial Vehicle Networks for Civil Applications: A Communications Viewpoint

The use of unmanned aerial vehicles (UAVs) is growing rapidly across many civil application domains, including real-time monitoring, providing wireless coverage, remote sensing, search and rescue, delivery of goods, security and surveillance, precision agriculture, and civil infrastructure inspection. Smart UAVs are the next big revolution in the UAV technology promising to provide new opportunities in different applications, especially in civil infrastructure in terms of reduced risks and lower cost. Civil infrastructure is expected to dominate more than $45 Billion market value of UAV usage. In this paper, we present UAV civil applications and their challenges. We also discuss the current research trends and provide future insights for potential UAV uses. Furthermore, we present the key challenges for UAV civil applications, including charging challenges, collision avoidance and swarming challenges, and networking and security-related challenges. Based on our review of the recent literature, we discuss open research challenges and draw high-level insights on how these challenges might be approached.

Unmanned Aerial Vehicles: A Survey on Civil Applications and Key Research Challenges

Deep reinforcement learning has emerged as a promising and powerful technique for automatically acquiring control policies that can process raw sensory inputs, such as images, and perform complex behaviors. However, extending deep RL to real-world robotic tasks has proven challenging, particularly in safety-critical domains such as autonomous flight, where a trial-and-error learning process is often impractical. In this paper, we explore the following question: can we train vision-based navigation policies entirely in simulation, and then transfer them into the real world to achieve real-world flight without a single real training image? We propose a learning method that we call CAD$^2$RL, which can be used to perform collision-free indoor flight in the real world while being trained entirely on 3D CAD models. Our method uses single RGB images from a monocular camera, without needing to explicitly reconstruct the 3D geometry of the environment or perform explicit motion planning. Our learned collision avoidance policy is represented by a deep convolutional neural network that directly processes raw monocular images and outputs velocity commands. This policy is trained entirely on simulated images, with a Monte Carlo policy evaluation algorithm that directly optimizes the network's ability to produce collision-free flight. By highly randomizing the rendering settings for our simulated training set, we show that we can train a policy that generalizes to the real world, without requiring the simulator to be particularly realistic or high-fidelity. We evaluate our method by flying a real quadrotor through indoor environments, and further evaluate the design choices in our simulator through a series of ablation studies on depth prediction. For supplementary video see: this https URL

CAD2RL: Real Single-Image Flight without a Single Real Image

Model predictive control of three-axis gimbal system mounted on UAV for real-time target tracking under external disturbances

In this paper, the concept of divergent component of motion (DCM, also called “Capture Point”) is extended to 3-D. We introduce the “Enhanced Centroidal Moment Pivot point” (eCMP) and the “Virtual Repellent Point” (VRP), which allow for the encoding of both direction and magnitude of the external forces and the total force (i.e., external plus gravitational forces) acting on the robot. Based on eCMP, VRP, and DCM, we present methods for real-time planning and tracking control of DCM trajectories in 3-D. The basic DCM trajectory generator is extended to produce continuous leg force profiles and to facilitate the use of toe-off motion during double support. The robustness of the proposed control framework is thoroughly examined, and its capabilities are verified both in simulations and experiments.

Three-Dimensional Bipedal Walking Control Based on Divergent Component of Motion

Urban search and rescue missions raise special requirements on robotic systems. Small aerial systems provide essential support to human task forces in situation assessment and surveillance. As external infrastructure for navigation and communication is usually not available, robotic systems must be able to operate autonomously. A limited payload of small aerial systems poses a great challenge to the system design. The optimal tradeoff between flight performance, sensors, and computing resources has to be found. Communication to external computers cannot be guaranteed; therefore, all processing and decision making has to be done on board. In this article, we present an unmanned aircraft system design fulfilling these requirements. The components of our system are structured into groups to encapsulate their functionality and interfaces. We use both laser and stereo vision odometry to enable seamless indoor and outdoor navigation. The odometry is fused with an inertial measurement unit in an extended Kalman filter. Navigation is supported by a module that recognizes known objects in the environment. A distributed computation approach is adopted to address the computational requirements of the used algorithms. The capabilities of the system are validated in flight experiments, using a quadrotor.

/pdf/toward-a-fully-autonomous-uav-research-platform-for-indoor-1561t1mp5j.pdf

Toward a Fully Autonomous UAV: Research Platform for Indoor and Outdoor Urban Search and Rescue

This paper gives an overview on the torque-controlled humanoid robot TORO, which has evolved from the former DLR Biped. In particular, we describe its mechanical design and dimensioning, its sensors, electronics and computer hardware. Additionally, we give a short introduction to the walking and multi-contact balancing strategies used for TORO.

/pdf/overview-of-the-torque-controlled-humanoid-robot-toro-22qgecw0kj.pdf

Overview of the torque-controlled humanoid robot TORO

We introduce our new quadrotor platform for realizing autonomous navigation in unknown indoor/outdoor environments. Autonomous waypoint navigation, obstacle avoidance and flight control is implemented on-board. The system does not require a special environment, artificial markers or an external reference system. We developed a monolithic, mechanically damped perception unit which is equipped with a stereo camera pair, an Inertial Measurement Unit (IMU), two processor-and an FPGA board. Stereo images are processed on the FPGA by the Semi-Global Matching algorithm. Keyframe-based stereo odometry is fused with IMU data compensating for time delays that are induced by the vision pipeline. The system state estimate is used for control and on-board 3D mapping. An operator can set waypoints in the map, while the quadrotor autonomously plans its path avoiding obstacles. We show experiments with the quadrotor flying from inside a building to the outside and vice versa, traversing a window and a door respectively. A video of the experiments is part of this work. To the best of our knowledge, this is the first autonomously flying system with complete on-board processing that performs waypoint navigation with obstacle avoidance in geometrically unconstrained, complex indoor/outdoor environments.

http://vigir.missouri.edu/~gdesouza/Research/Conference_CDs/IEEE_IROS_2013/media/files/0927.pdf

Stereo vision based indoor/outdoor navigation for flying robots

Micro air vehicles have become very popular in recent years. Autonomous navigation of such systems plays an important role in many industrial applications as well as in search-and-rescue scenarios. We present a quadrotor that performs autonomous navigation in complex indoor and outdoor environments. An operator selects target positions in the onboard map and the system autonomously plans an obstacle-free path and flies to these locations. An onboard stereo camera and inertial measurement unit are the only sensors. The system is independent of external navigation aids such as GPS. No assumptions are made about the structure of the unknown environment. All navigation tasks are implemented onboard the system. A wireless connection is only used for sending images and a three-dimensional 3D map to the operator and to receive target locations. We discuss the hardware and software setup of the system in detail. Highlights of the implementation are the field-programmable-gate-array-based dense stereo matching of 0.5 Mpixel images at a rate of 14.6 Hz using semiglobal matching, locally drift-free visual odometry with key frames, and sensor data fusion with compensation of measurement delays of 220i¾?ms. We show the robustness of the approach in simulations and experiments with ground truth. We present the results of a complex, autonomous indoor/outdoor flight and the exploration of a coal mine with obstacle avoidance and 3D mapping.

Autonomous Vision-based Micro Air Vehicle for Indoor and Outdoor Navigation

Multi-view stereo algorithms are an attractive technique for the digital reconstruction of outdoor sites. Concerning the data acquisition process a vertical take off and landing UAV carrying a digital camera is a suitable platform in terms of mobility and flexibility in viewpoint placement. We introduce an automated UAV based data acquisition and outdoor site reconstruction system. A special focus is set on the problem of model based view planning using a coarse digital surface model (DSM) with minimal data preprocessing. The developed view planning heuristic considers a coverage, a maximum view angle and an overlapping constraint imposed by multi-view stereo reconstruction techniques. The time complexity of the algorithm is linear with respect to the size of the area of interest. We demonstrate the efficiency of the entire system in two scenarios, a building and a hillside.

Korbinian Schmid

Papers

Toward a Fully Autonomous UAV: Research Platform for Indoor and Outdoor Urban Search and Rescue

Overview of the torque-controlled humanoid robot TORO

Stereo vision based indoor/outdoor navigation for flying robots

Autonomous Vision-based Micro Air Vehicle for Indoor and Outdoor Navigation

View Planning for Multi-View Stereo 3D Reconstruction Using an Autonomous Multicopter