Sharif Digital Repository

Tight unmanned aerial vehicle (UAV) autonomous missions such as formation flight (FF) and aerial refueling (AR) require an active controller that works in conjunction with a precise sensor that is able to identify an in-front aircraft and to estimate its relative position and orientation. Among possible choices vision sensors are of interest because they are passive in nature and do not require the cooperation of the in-front aircraft in any way. In this paper new vision-based estimation and navigation algorithms based on neural networks is developed. The accuracy and robustness of the proposed algorithm have been validated via a detailed modeling and a complete virtual environment based on... 

In this paper, a new nonlinear robust adaptive impedance controller is addressed for Unmanned Aerial Vehicles (UAVs) equipped with a robot manipulator that physically interacts with environment. A UAV equipped with a robot manipulator is a novel system that can perform different tasks instead of human being in dangerous and/or inaccessible environments. The objective of the proposed robust adaptive controller is control of the UAV and its robotic manipulators end-effector impedance in Cartesian space in order to have a stable physical interaction with environment. The proposed controller is robust against parametric uncertainties in the nonlinear dynamics model of the UAV and the robot... 

Design of a vision-based target tracking control strategy for a dual-rotor UAV with tilting rotors is presented. In this research, the under-study UAV is equipped with a sliding mass that can slide over a motorized linear track attached to the UAV's fuselage enabling it to perform sophisticated maneuvers by monitoring its five controllable degrees of freedom. The dynamic model of the system is first derived and then a model-based controller using vision based pose estimation, obtained via an onboard down-looking camera, is designed enabling the UAV to: (1) pitch hover in position, (2) servo towards a target position, and (3) track a moving target. Simulations show the premise of the proposed... 

The use of Model Aided Inertial Navigation (MAIN) during the landing of an Unmanned Aerial Vehicle (UAV) is investigated. A new MAIN algorithm is proposed, which is fast and accurate enough to be used in automatic landing. In this algorithm, the six Degree of Freedom (6DoF) model of the UAV is tightly coupled with the inertial navigation system; thus, the 6DoF model acts as an aiding system for the INS and vice versa. In the last parts of the landing phase in proximity of Earth, the proposed algorithm also estimates and removes the Ground Effect (GE) uncertainties and provides the height controller with a realistic model. An adaptive controller based on a parametric state-space model is used... 

In this paper, the development of a controller for autonomous lateral alignment of fixed-wing Unmanned Aerial Vehicles (UAVs) with runway centerline in landing phase is presented. Fuzzy Logic Control (FLC) is used in order to enable the vehicle to mimic the decision making procedure that a pilot follow in the same situation. Also, for longitudinal motion controller design for the UAV to follow a pre-defined trajectory, the pole-placement technique is used. It is assumed that the runway relative position and orientation are provided by a built in vision system and its associated image processing unit. The performance of the controller in the presence of the Gaussian noises is investigated by... 

Quadrotors have become very popular, both commercially and as research platforms. Numerous studies have been carried out on these aerial vehicles, covering different aspects of their dynamics and control. Many of the interesting applications of quadrotors arise when they are used as part of a cooperative robotic system, consisting of aerial and ground robots. Two essential steps toward realizing autonomous teams of aerial and ground robots are achieving position coordination between them, and autonomous landing of aerial robots on ground platforms. The goal of this paper is to tackle the problems of tracking and landing of an aerial robot on a mobile platform. Two approaches are presented... 

A cooperative task allocation and search algorithm is proposed to find and localize a group of ground-based moving targets using a group of Unmanned Air Vehicles (UAVs) working in a decentralized manner. It is assumed that targets have RF emissions. By using an algorithm including Global Search (GS), Approach Target (AT), Locate Target (LT), and Target Reacquisition (TR) modes, UAVs cooperatively search the entire parts of a desired area, approach to the detected targets, locate the targets, and search again to find the targets that stop transmitting their RF emissions during the localization process, respectively. In the GS mode, UAVs utilize a cost function to select the best zone for... 

A cooperative task allocation and search algorithm is proposed to find and localize a group of ground-based moving targets using a group of Unmanned Air Vehicles (UAVs) working in a decentralized manner. It is assumed that targets have RF emissions. By using an algorithm including Global Search (GS), Approach Target (AT), Locate Target (LT), and Target Reacquisition (TR) modes, UAVs cooperatively search the entire parts of a desired area, approach to the detected targets, locate the targets, and search again to find the targets that stop transmitting their RF emissions during the localization process, respectively. In the GS mode, UAVs utilize a cost function to select the best zone for... 

In recent years, unmanned aircraft systems (UASs) are frequently used in many different applications of photogrammetry such as building damage monitoring, archaeological mapping and vegetation monitoring. In this paper, a new vertical take-off and landing fixed-wing UAS is proposed to robust photogrammetry missions, called SAMA-VTOL. In this study, the capability of SAMA-VTOL is investigated for generating orthophoto. The major stages are including designing, building and experimental scenario. First, a brief description of design and build is introduced. Next, an experiment was done to generate accurate orthophoto with minimum ground control points requirements. The processing step, which... 

A novel fault-tolerant model predictive control (MPC)-based trajectory tracking approach for an aerial vehicle is presented in this study. A generalised online sequential extreme learning machine is introduced first to identify the corresponding coefficients of actuator faults. Subsequently, a robust trajectory tracking control is developed based on MPC, where the system constraints can be effectively considered in the designed control scheme. Trajectory tracking control is achieved by controlling only the acceleration of the aerial robot in the MPC structure. This leads to less computational burden and faster closed-loop dynamics. In addition, an effective disturbance observer is employed,... 

In recent years, unmanned aerial systems have attracted great attention due to the electronic systems technology advancements. Among these vehicles, unmanned helicopters are more important because of their special abilities and superior performance. The complex nonlinear dynamic system (caused by main rotor flapping dynamics coupled with the rigid body rotational motion) and considerable effects of ambient disturbance make their utilization hard in actual missions. Attitude dynamics have the main role in helicopter stabilization, so implementing proper control system for attitude is an important issue for unmanned helicopter hovering and trajectory tracking performance. Besides this,... 

In this paper, a new three-dimensional path planning approach with obstacle avoidance for UAVs is proposed. The aim is to provide a computationally-fast on-board sub-optimal solution for collision-free path planning in static environments. The optimal 3D path is an NP (non-deterministic polynomial-time) hard problem which may be solved numerically by global optimization algorithms such as the Particle Swarm Optimization (PSO). Application of PSO to the 3D path planning class of problems faces typical challenges such slow convergence rate. It is shown that the performance may be improved markedly by implementing a novel parallel approach and incorporation of new termination conditions.... 

In this study, a new heuristic multiple model filter, called Multiple Model Extended Continuous Ant Colony Filter, is proposed to solve a nonlinear multiple model state estimation problem. In this filter, a bank of extended continuous ant colony filters are run in parallel to solve the multiple model estimation problem. The probability of each model is continually updated and consequently both the true model and the states of the nonlinear system are updated based on the weighted sum of the filters. The new multiple model filter is tested on an engineering problem. The problem is to estimate simultaneously the states of a fixed-wing unmanned aerial vehicle as well as the wind model, applied... 

Visibility is an important topic in computer graphics, motion planning, and computational geometry. To deal with the increasing complexity of the scenes considered, some research has been performed in visibility processing in order to accelerate the visibility determination. Two of the most studied such structures are visibility graph and visibility complex. Visibility graph is a fundamental geometric structure which is used in many applications, including illumination and rendering, motion planning, pattern recognition, and sensor networks. While the concept of visibility graph is widely studied for 2D scenes, there is no acceptable equivalence of visibility graph for 3D space. Similarly,... 

Sensor networks are systems of many small units that work together to monitor a given environment. A mixed sensor network is composed of a collection of static and mobile nodes that has sensing, computation, communication, and locomotion capabilities. Sensor mobility can be used to compensate for the lack of sensors and improve network coverage. This paper aims to describe an algorithm called Coin for mixed sensor network deployment in an unknown environment by using a small number of mobile sensors. It allows the network to achieve maximum coverage with minimum energy consumption. Our solution includes distributed schemes for: 1) determining the exact size of a coverage hole; 2) determining... 

Single-task learning in artificial neural networks will be able to learn the model very well, and the benefits brought by transferring knowledge thus become limited. In this regard, when the number of tasks increases (e.g., semantic segmentation, panoptic segmentation, monocular depth estimation, and 3D point cloud), duplicate information may exist across tasks, and the improvement becomes less significant. Multi-task learning has emerged as a solution to knowledge-transfer issues and is an approach to scene understanding which involves multiple related tasks each with potentially limited training data. Multi-task learning improves generalization by leveraging the domain-specific information... 

The combination of pursuit and line of sight guidance laws, called PLOS, is used to steer an unmanned aerial vehicle along a desired path. In the previous studies, the parameters of this guidance law are tuned by trial and error and are constant, during the flight. In this research, it will be shown that the optimal value of these parameters depends on the initial conditions of the problem and the wind conditions. For this reason, a fuzzy system is proposed to generate the instantaneous optimal value of these parameters, in such a way that the flying vehicle converges to the desired path in less time and follows it more accurately, in the presence of wind. For this purpose, a cost function... 

This paper presents an efficient mixed-integer linear formulation for long term maintenance schedule of overhead lines. The proposed formulation is based on risk management approach and utilizes the model of decoupled risk factors. The proposed methodology yields a significant computational saving comparing to the previously reported dynamic programming. Risk management approach enables the asset managers to consider the actual condition of electrical equipment and expected consequence of their failures. Furthermore, the decoupled risk strategy in conjunction with the mixed-integer linear programming formulation establishes a precise description of time-dependent deterioration failure rate...