Guray Yilmaz

Establishing Obstacle and Collision Free Communication Relay for UAVs with Artificial Potential Fields

In this paper, Unmanned Aerial Vehicles are used for establishing an airborne communication relay chain to extend the communication range or to obtain a channel between two far points which are outside the single UAV communication range. Positions of the UAVs in the chain are detected by the vehicles autonomously and while establishing the suitable formation, collision avoidance between vehicles and other geographical obstacles are considered by using artificial potential fields. Especially to provide reliable continues communication between vehicles as uninterrupted channels, positions of the UAVs which are providing line of sight, calculated automatically by tuning artificial potential field parameters dynamically. The success of this novel approach is expressed by simulation studies in Matlab environment and the simulation results are validated using NS2 simulator.

Real-time Autonomous UAV Formation Flight with Collision and Obstacle Avoidance in Unknown Environment

In this paper autonomous formation control for Unmanned Aerial Vehicles (UAVs) has been discussed and a real time solution has been put forward by benefiting General Purpose Graphical Processing Units (GPGPU) accelerated potential field approach while ensuring obstacle and collision avoidance in unknown environment by using real-time sensors. GPGPU accelerated real time formation control for UAVs was designed and the basic model of the approach has been explained in our previous work (Cetin and Yilmaz 2014). As the deficiencies of the previous approach, autonomous real time collision and coordinated obstacle avoidance features in unknown environments are also handled while maintaining formation flight conditions in this work. With these features, improved autonomous formation control approach is discussed as a real time solution. The computation is performed by using Graphical Processing Units (GPUs) as parallel computation architectures by benefiting from Single Instruction Multiple Data (SIMD) type parallel algorithms. Classic binary map conversation, connected component labeling and minimum bounding box algorithms which are commonly used for image processing applications, has been evaluated for real time obstacle detection and avoidance features by developing GPGPU suitable parallel algorithms. Real-time solution has been developed by integrated these parallel algorithms with parallel Artificial Potential Field (APF) computation algorithm. Simulation results are proved that this novel autonomous improved formation control approach is successful and it would be used in real time applications like UAV formation flight missions.

A novel fast and accurate algorithm for Terrain Referenced UAV localization

Recently, unmanned aerial vehicles (UAVs) have become one of the most popular and promising means for both military and civilian posts and academic research areas. Localization of the UAVs and persistent tracking of a UAV have vital importance to provide a UAV with navigation information and help to cope with getting lost permanently. Indeed, Inertial Navigation System (INS) and Global Positioning System (GPS) seem to be adequate for navigation of UAVs. However, an alternative augmented navigation system for UAVs should be taken into consideration since INS has accumulated errors and GPS always has the possibility of jamming and satellite signal loss. Terrain Referenced Navigation (TRN) could be a good alternative as a decision support system for these main systems. This study aims to detect the location of a lost or GPS-disabled UAV throughout a planned flight by using only the terrain data. In addition, assumptions and limitations are minimized for the sake of simplifying the process to apply this methodology on a real UAV in the future, e.g. flight through all directions with physically possible turn rates is allowed. In order to provide data of the terrain, Digital Elevation Model (DEM) of the flight region with 30m resolution is exploited. The proposed method is based on searching and matching the collected elevation values of the terrain below UAV within the DEM and makes use of simulation techniques to test the accuracy and performance. The whole algorithm utilizes a sequence of elevation values with a predefined length (i.e. profile). Mainly, all possible profiles are generated before the flight and stored in a huge search space. We identify, sort and classify these elevation profiles in order to perform search operations in a small subset of the huge set. During the flight, a sequence of terrain elevations, which is computed with the help of radar and barometric altimeter measurements, is searched within a small neighborhood of corresponding profile set.

2D path planning for UAVs in radar threatening environment using simulated annealing algorithm

Path planning for an Unmanned Aerial Vehicle which can be used in many different purposes is a challenging issue especially in tasks involving large number of visiting points. The problem becomes more complex when the flight environment is taken into account with numerous constraints. In this paper, Simulated Annealing (SA) algorithm is used to obtain nearly optimal path in 2D radar constrained environment. A simple threat avoidance approach is developed and applied to the solution found by using SA in order to escape from regular circular radar threats. Tests are conducted to observe the behaviour of SA algorithm and proposed threat avoidance approach. The results show that, in a reasonable period of time, SA algorithm provides acceptable solutions with its capability of escaping from local minima using Metropolis acceptance rule and the proposed threat avoidance approach applied to the best found solution makes the path threat-free simply.

A Terrain Referenced UAV Localization Algorithm Using Binary Search Method

This study focuses on localization of Unmanned Aerial Vehicles (UAV) since permanent navigation has vital significance to support position information and to avoid getting lost. Actually, there exist effective aeronautical navigation systems in use. Inertial Navigation System (INS) and Global Positioning System (GPS) are two representatives of the most common systems utilized in traditional aerial vehicles. However, an alternative supporter system for UAVs should be mentioned since INS and GPS have serious deficiencies for UAVs such as accumulated errors and satellite signal loss, respectively. Such handicaps are coped with integrating these systems or exploiting other localization systems. Terrain Referenced Navigation (TRN) could be a good alternative as a supporter mechanism for these main systems. This study aims to localize a UAV accurately by using only the elevation data of the territory in order to simulate a TRN system. Application of the methodology on a real UAV is also considered for the future. Thus assumptions and limitations are designed regarding the constraints of real systems. In order to represent terrain data, Digital Elevation Model (DEM) with original 30 meter-resolution (Eroglu and Yilmaz 2013) and also synthetically generated 10 meter-resolution maps are utilized. The proposed method is based on searching the measured elevation values of the flight within the DEM and makes use of simulation techniques to test the accuracy and the performance. The whole system uses sequences of elevation values with a predefined length (i.e. profile). Mainly, all possible profiles are generated and stored before the flight. We identify, classify and sort profiles to perform search operations in a small subset of the terrain. During the flight, a measured flight profile is searched by the Binary search method (Eroglu 2013) within a small neighborhood of corresponding profile set.

Development of a GPU Accelerated Terrain Referenced UAV Localization and Navigation Algorithm

This study focuses on localization and navigation of Unmanned Air Vehicles (UAVs) based on digital terrain map data. The solution to the Terrain Referenced Localization and Navigation (TERELONA) or Terrain Referenced Navigation (TRN) is described by using particle filter. In many UAV applications one of the most important points is to provide accurate location information continuously. TERELONA system can supply the air vehicle with the accurate position information with a bounded error. In this paper, the particle filtering method as an implementation of Bayesian approach to the terrain referenced localization and navigation is described. The radar altimeter measurements are used as an implicit representation of aircraft position. Whenever new measurements are taken from radar altimeter, they are compared to the Digital Terrain Map (DTM) data in order to fix a position. The solution is represented, in a Bayesian framework, by a set of particles with their corresponding weights. We have developed the terrain referenced localization and navigation algorithm based on the particle approximation. The proposed algorithm, which is developed in CUDATM, is also tested on the GPU environment using GPUmat software architecture. Thus, we can cope with the computational load of the very large initial horizontal position errors. The proposed algorithm has been implemented in MATLABTM environment and evaluated on simulated data. Simulations are conducted over an ASTER GDEM product which belongs to a region in northwest of Turkey. The simulation results are provided.

Sigmoid Limiting Functions and Potential Field Based Autonomous Air Refueling Path Planning for UAVs

In this paper autonomous air-refueling (AAR) path planning for Unmanned Aerial Vehicles (UAVs) has been discussed and an enhanced approach has been put forward. AAR path planning for UAVs was designed and the basic model of the pattern was put forward in our previous work (Cetin and Yilmaz 2013). Additionally to our previous works, the deficiencies of the previous approach, like smooth maneuvers in the tanker approach and the boundary functions of the potential zones has been handled, furthermore special pattern parameters are added to the approach which makes it suitable for different kind of UAVs that has variable flight speed and turn radius parameters. An important originality of the approach is using of sigmoid limiting functions while modeling dynamic behaviors of the potential fields that are based on path planning algorithms. In order to use the AAR path planning approach in a real time application, the computation is performed in Graphical Processing Units (GPUs) based parallel architecture by benefiting from many cores in General Purpose Graphical Processing Units (GPGPU) as described in previous research (Cetin and Yilmaz 2013). With the addition of the sigmoid limiting functions instead of logical binary boundary functions computation needs of the autonomous approach become higher point and the only way to use the approach in the real time applications is benefiting of the parallel computing approach. The comparison of the boundary functions as computational performance and path outputs are discussed with the simulation results in this paper. Simulation results are proved that this novel autonomous parallel path planning approach is successful and it would be used in real time applications like AAR mission.

Wireless sensor networks for space and Solar-system missions

Wireless sensor networks (WSNs) are multi-hop self-organizing networks which include a huge number of nodes integrating environmental measuring, data processing and wireless communications in order to apprehend, collect and process information to achieve defined tasks. A diverse set of applications for WSNs encompassing different fields have already emerged including environmental applications, inventory monitoring, military applications, intrusion detection, health applications, motion tracking, machine malfunction detection and etc. Among these application areas the use of WSNs can adapted to Space and Solar-system missions. In the last years, space-based WSNs have gained increasing attention from both the research communities and companies involved in space research. This paper outlines the usage of a space-based wireless sensor networks (SB-WSNs), which applies the concept of terrestrial wireless sensor networks to the space.

GIS applications in cloud computing platform and recent advances

There is a rising requirement for high quality satellite photos and also fast distribution of this kind of spatial data. Geographic Information Systems (GIS) platforms are needed to be implemented with the capability of dealing with the massive spatial data, especially image data. Thus these platforms need computational power and high capacity data storage. Nowadays Cloud Computing is one of the most popular IT solutions for GIS. It serves tremendous computing resources over the Internet. Users do not need to know neither the location of the servers which the applications run on nor the data storages their information is collected. The rise of Cloud Computing is making notable change in various areas of information technology. GIS technology users are one of the first customers to explore the cloud-based applications and its benefits. Users take the advantage of sharing spatial data and applications using the cloud. Therefore massive spatial data storages fit into cloud instead of transferring the data via the network.

A new method for anomaly detection and target recognition

Use of unmanned Aerial Vehicles (UAVs) has gained significant importance in the recent years because they are capable of to be used in in civilian and military purposes for reconnaissance, surveillance, disaster relief, among other tasks. In this paper we present new automated anomaly detection and target recognition methodology that can be used on such a UAV. The standard paradigm for anomaly detection and target recognition in hyperspectral imagery (HSI) is to run a detection or recognition algorithm, typically statistical in nature, and visually inspect each high-scoring pixel to decide whether it is an anomaly or background data. A new method of anomaly detection and target recognition in HSI was studied based on a Neural Network (NN). Two multi-layered neural networks are used for anomaly detection and target recognition. The first phase of the model is used to detect anomalies in HSI. The second phase of the model is to use determine whether the anomaly is a predefined target or not. Both networks are trained in accordance with its intended purpose, so increase in performance is provided. This method can be a suitable solution for applications where the unmanned aerial vehicles used.