Micro Unmanned Aerial Vehicles Research Articles

Cognitive Agent Mobility for Aerial Sensor Networks

For efficient sensor coverage of large industrial and incident areas, fast and flexible strategies for collecting sensor data through an autonomous, wirelessly connected swarm of (Micro) Unmanned Aerial Vehicles (MUAVs) are still an emerging challenge. Deploying multiple MUAVs which stably carry sensing equipment in hostile environments yields cost efficiency and reducing the risk to human life. The use of an aerial ad hoc sensor network based on MUAV agents promises more timely and accurate information by fusing measurements from different types of sensors. In this paper, we examine agent-based mobility algorithms that target high spatial coverage distribution, high total coverage, and high-quality communication links. For these key figures, the performance of the different novel mobility algorithms is evaluated. We particularly focus on channel aware mobility and on self-organizing mesh topologies of MUAV-based sensor swarms with respect to communication constraints.

The GRASP Multiple Micro-UAV Testbed

In the last five years, advances in materials, electronics, sensors, and batteries have fueled a growth in the development of microunmanned aerial vehicles (MAVs) that are between 0.1 and 0.5 m in length and 0.1-0.5 kg in mass [1]. A few groups have built and analyzed MAVs in the 10-cm range [2], [3]. One of the smallest MAV is the Picoftyer with a 60-mmpropellor diameter and a mass of 3.3 g [4]. Platforms in the 50-cm range are more prevalent with several groups having built and flown systems of this size [5]-[7]. In fact, there are severalcommercially available radiocontrolled (PvC) helicopters and research-grade helicopters in this size range [8].

Wind-field reconstruction from flight data using an unbiased minimum-variance unscented filter

Although guidance of all aircraft is affected by wind disturbances, micro-unmanned aerial vehicles are especially susceptible. To estimate unknown wind disturbance, we consider two illustrative scenarios for planar flight. In the first scenario, we assume that measurements of the heading angle are available, while, in the second scenario, we assume that measurements of the heading angle are not available. Since the disturbance estimation problem is non-linear, we develop an extension of the unscented Kalman filter that provides an estimate of the unknown wind disturbance. Furthermore, we show through simulations that, when the heading angle is not measured, a kinematic ambiguity is introduced. However, when the initial heading angle is known and the subsequent heading angle is not measured, this kinematic ambiguity is resolved and accurate estimates of the wind velocity are obtained.

Mission Completion Time is Sensitive to Teleoperation Performance During Simulated Reconnaissance Missions with a Micro-Unmanned Aerial Vehicle

Mission Completion Time is Sensitive to Teleoperation Performance During Simulated Reconnaissance Missions with a Micro-Unmanned Aerial Vehicle

Real-Time Vision Sensor for an Autonomous Hovering Micro Unmanned Aerial Vehicle

Visionalgorithmswereimplementedonanfieldprogrammablegatearraytoprovideadditional information to supplement the insufficient data of a standard inertial measurement unit in order to create a previously unrealized completely onboard vision system for microunmanned aerial vehicles.The onboard vision system is composed of an field programmable gate array board, and a custom interface daughterboard which allow it to provide data regarding drifting movements of the micro-unmanned aerial vehicle not detected by inertial measurement units. The algorithms implemented for the vision system include a Harris feature detector, template matching feature correlator, similarity-constrained homography by random sample consensus, color segmentation, radial distortion correction, and an extended Kalman filter with a standard-deviation outlier rejection technique. This vision system was designed specifically for use as an onboard vision solution for determining movement of micro-unmanned aerial vehicles that have severe size, weight, and power limitations. Results show that the vision system is capable of real-time onboard image processing with sufficient accuracy to allow a micro-unmanned aerial vehicle to control itself without power or data tethers to a groundstation.

Mission Completion Time is Sensitive to Teleoperation Performance during Simulated Reconnaissance Missions with a Micro-Unmanned Aerial Vehicle

In prior research, we found that temporal measures were affected by input device (game controller vs. mouse); however, other measures were not (e.g., number of targets detected, or number of collisions). The main purpose of the present research was to investigate the sensitivity of a temporal performance measure for simulated reconnaissance missions. Instead of measuring number of targets photographed in a fixed amount of time, this study measured time to photograph a set number of targets. Twenty participants went through simulation-based micro-unmanned aerial vehicle operator training, including three skill courses and two reconnaissance missions. Half the participants used a mouse as their input device, whereas the rest used a game controller. Mission completion time was found to be faster when participants used the game controller than when they used the mouse, for both skill missions and reconnaissance missions. Effective simulation-based training for micro-unmanned vehicles operators requires both relevant simulated missions and objective performance feedback and mastery criteria. These results suggest that mission completion time could provide a good performance measure for this purpose.

Extending the Evolutionary Robotics approach to flying machines: An application to MAV teams

The work presented in this article focuses on the use of embodied neural networks--developed through Evolutionary Robotics and Multi-Agent Systems methodologies--as autonomous distributed controllers for Micro-unmanned Aerial Vehicle (MAV) teams. The main aim of the research is to extend the range of domains that could be successfully tackled by the Evolutionary Robotics approach. The flying robots realm is an area that has not been yet thoroughly investigated by this discipline. This is due to the lack of an affordable and reliable robotic platform to use for carrying out experiments, and to the difficulty and the high computational load involved in experiments based upon a realistic software simulator for aircraft. We believe that the most recent improvements to the state of the art now permit the investigation of this domain. For demonstrating this point, two different evolutionary computer simulation models are presented in this article. The first model, which uses a simplified 2D test environment, has resulted in controllers evolved with the following capabilities: (1) navigation through unknown environments, (2) obstacle-avoidance, (3) tracking of a movable target, and (4) execution of cooperative and coordinated behaviors based on implicit communication strategies. In order to improve the robustness of these results and their potential use in real MAV teams, a more sophisticated 3D model is presented herein. The results obtained so far using the two models demonstrate the feasibility of the chosen approach for further research on the design of autonomous controllers for MAVs.

The Effects of Input Device and Latency on Ability to Effectively Pilot a Simulated Micro-UAV

The effects of input device and latency during training to manually pilot a simulated micro-unmanned aerial vehicle (MAV) were investigated. In our prior simulation research, MAV missions were completed more quickly when trainees used a game controller than when they used a mouse. This experiment investigated whether this difference would persist when participants were tested in a novel environment and when some realistic latency was imposed between input command and MAV response. Fifty-six participants were trained to operate a MAV in one simulated environment and then tested in a novel environment. Four between-group conditions were examined, formed by crossing two 2-level factors: input device (mouse vs. game controller) and latency period (no time delay vs. 500 ms delay). The effects of input device replicated our prior research and also transferred to the novel environment; no substantial effects of delay were found.

ROBUST FLATNESS BASED CONTROL AND MOTION PLANNING OF A MUAV

The design of a controller combining flatness and sliding modes has been chosen for a Micro Unmanned Aerial Vehicle (MUAV), a highly non linear system. The stabilization of the MUAV is known to be not so simple when it must ensure complex trajectories tracking in a given specified mission. Thus, the controller, to be designed, needs to achieve an asymptotic stabilization with a certain level of robustness, particularly in a turbulent environment. We will show that the proposed model, the choice of state variables, and the controller coupled with Lagragian interpolation of the “stop-stop” trajectories, leads to satisfactory results, in terms of robustness with regards to structured uncertainties and exogenous disturbances. Copyright © 2007, IFAC.

Human Factors and Trainability of Piloting a Simulated Micro Unmanned Aerial Vehicle

In 2004, the U.S. Army Research Institute's (ARI) Simulator Systems Research Unit began studies involving the training requirements for operators of a micro-unmanned aerial vehicle (MAV). Our research involved the use of a touch-screen operator control interface developed for the DARPA MAV Advanced Technology Demonstration. This control system allowed operators to plan and run autonomous flight missions or to tele-operate a simulated MAV around a static synthetic environment. An initial study focused primarily on the usability of the system. Extensive heuristic evaluations were conducted by seven volunteers with backgrounds in human factors and military training systems. Each evaluator completed a self-paced training session including exercises that tested their ability to perform various control functions. Lack of immediate feedback from touch-screen inputs and missing or obscure status information formed the basis of several of the usability issues. Manually piloting the MAV presented the most difficulty to operators. As such, a second study was conducted that focused specifically on manual control tasks. In this study, participants were trained on manual control of the MAV, and then completed four increasingly difficult missions, requiring them to pilot the vehicle through the synthetic environment. This experiment was designed to compare the effect of supplemental sensor imagery on mission performance. During Study 1, operators could choose to view a sensor image taken from a fixed camera pointed 15 degrees below horizontal or straight down (90 degrees), but only one view was available at a time. During Study 2, operators were provided with three sensor views simultaneously. The 15-degree view was presented in a primary sensor window, and two additional views were displayed in smaller windows below it. The camera angle of one of these supplemental views was the manipulated independent variable — 30, 60, or 90 degrees from horizontal. The remaining window always contained an overhead satellite view (downward view from 5000 feet above the MAV). Data were collected on time to complete each mission, the number of physical interactions each user made with the interface, SME ratings, workload, and usability. Results indicated that mission requirements had a greater effect on performance and workload ratings than the camera angle of the supplemental view, though the camera angle of the supplemental view did affect mission time required to capture images of designated target buildings. Session averages of workload, usability, mission completion time, and SME rating were significantly inter-correlated. Higher SME ratings were associated with lower participant ratings of workload, shorter mission completion times, and higher usability ratings.