Keith W. Sevcik

Autonomous Landing for Indoor Flying Robots Using Optic Flow

Urban environments are time consuming, labor intensive and possibly dangerous to safe guard. Accomplishing tasks like bomb detection, search-andrescue and reconnaissance with aerial robots could save resources. This paper describes a prototype called CQAR: Closed Quarter Aerial Robot, which is capable of flying in and around buildings. The prototype was analytically designed to fly safely and slowly. An optic flow microsensor for depth perception, which will allow autonomous takeoff and landing and collision avoidance, is also described.

A competition to identify key challenges for unmanned aerial robots in near-earth environments

Tasks like bomb-detection, search-and-rescue, and reconnaissance in near-Earth environments are time, cost and labor intensive. Aerial robots could assist in such missions and offset the demand in resources and personnel. However, flying in environments rich with obstacles presents many more challenges which have yet to be identified. For example, telephone wire is one obstacle that is known to be hard to detect in mid-flight. This paper describes a safe and easy to fly platform in conjunction with an aerial robot competition to highlight key challenges when flying in near-Earth environments

Exploring search-and-rescue in near-earth environments for aerial robots

Homeland security missions executed in near-earth environments are often time consuming, labor intensive and possibly dangerous. Aerial robots performing tasks such as bomb detection, search-and-rescue and reconnaissance could be used to conserve resources and minimize risk to personnel. Flying in environments which are heavily populated with obstacles yields many challenges. Little data exists to guide the design of vehicles and sensor suites operating in these environments. This paper explores the challenges encountered implementing several different sensing technologies in near-earth environments. The results of applying these technologies to control a robotic blimp are presented to direct future work

Exploring the Effect of Obscurants on Safe Landing Zone Identification

Manned rotorcraft are often employed in harsh environments and difficult terrain that are inaccessible to other craft. Conversely, robotic rotorcraft are operated in controlled settings, often at safe, high altitudes. Missions such as cargo delivery, medevac and fire fighting are unachievable because of unpredictable adverse environmental conditions. To enable UAVs to perform these missions, the effects of obscurants on UAV sensor suites and algorithms must be clearly understood. This paper explores the use of a laser range finder to accomplish landing zone identification in unknown, unstructured environments. The ability to detect a landing zone in environments obscured by smoke is investigated. This is accomplished using a design methodology of testing and evaluating in a controlled environment followed by verification and validation in the field. This methodology establishes a concrete understanding of the sensor performance, thereby removing ambiguities in field tests.

Development and Evaluation of a Chase View for UAV Operations in Cluttered Environments

Civilian applications for UAVs will bring these vehicles into low flying areas cluttered with obstacles such as building, trees, power lines, and more importantly civilians. The high accident rate of UAVs means that civilian use will come at a huge risk unless we design systems and protocols that can prevent UAV accidents, better train operators and augment pilot performance. This paper presents two methods for generating a chase view to the pilot for UAV operations in cluttered environments. The chase view gives the operator a virtual view from behind the UAV during flight. This is done by generating a virtual representation of the vehicle and surrounding environment while integrating it with the real-time onboard camera images. Method I presents a real-time mapping approach toward generating the surrounding environment and Method II uses a prior model of the operating environment. Experimental results are presented from tests where subjects flew in a H0 scale environment using a 6 DOF gantry system. Results showed that the chase view improved UAV operator performance over using the traditional onboard camera view.

Improving unmanned aerial vehicle pilot training and operation for flying in cluttered environments

Future applications will bring unmanned aerial vehicles (UAVs) to new environments such as urban areas, causing a change in the way that UAVs are currently operated. However, UAV accidents still occur at a much higher rate than the accident rate for commercial airliners. Therefore, there is a need to better train UAV pilots and augment their performance to minimize accidents. In this paper, the authors present two methods for generating a chase view point (similar to the view of being towed behind the aircraft). Through use of these viewpoints, the authors propose to increase the situational awareness of UAV operators when flying in cluttered environments. The chase view combines a rotated onboard camera view with a virtual representation of the vehicle and the surrounding operating environment. Experiments were conducted evaluating a chase view versus a traditional onboard camera view during UAV flights using a 6 DOF gantry system. Results showed that the chase view improved UAV operator performance.

Testing Unmanned Aerial Vehicle Missions in a Scaled Environment

UAV research generally follows a path from computer simulation and lab tests of individual components to full integrated testing in the field. Since realistic environments are difficult to simulate, its hard to predict how control algorithms will react to real world conditions such as varied lighting, weather, and obstacles like trees and wires. This paper introduces a methodic approach to developing UAV missions. A scaled down urban environment provides a facility to perform testing and evaluation (T&E) on control algorithms before flight. A UAV platform and test site allow the tuned control algorithms to be verified and validated (V&V) in real world flights. The resulting design methodology reduces risk in the development of UAV missions.

Designing aerial robot sensor suites to account for obscurants

Recent events have proven the utility of aerial robots in search and rescue and military operations performed in near-Earth environments. However, autonomy in such environments is still limited by the robots ability to perform collision avoidance. Real time, high fidelity maps can be utilized to assist the robot in performing collision avoidance. This paper explores the use of a laser range finder to accomplish real time map building in the presence of obscurants. Experiments were conducted to determine the probabilistic models of the sensor in the presence of rain and fog. These models are then utilized to map obstacles and terrain features in the presence of obscurants. Preliminary results show the implementation of these methods in an aerial vehicle testing facility.

Towards Scaled Designing and Testing of Unmanned Aerial Vehicle Missions

Today’s UAVs are being tasked to fly missions in increasingly difficult environments. Buildings, trees and thin wires form challenging terrain for the UAV to negotiate. The current paradigm of UAV research typically moves from computer simulation and individual subsystem testing to full integration in the field. The reactions of control algorithms to realistic sensor data are difficult to capture in simulation and can result in costly crashes. This paper introduces a methodic approach to developing UAV missions. A scaled down urban environment provides a facility to perform testing and evaluation (T&E) on control algorithms before flight. A UAV platform and test site allow the tuned control algorithms to be verified and validated (V&V) in real world flights. The resulting design methodology reduces risk in the development of UAV missions.Copyright