researches have been carried out to develop such a NGC system for Springer with main work been summarised in Table 1. It can be found that most of these studies have been focused on the development of guidance and control modules in the NGC system. For example, interval Kalman filtering as well as the fuzzy-logic based multi-sensor data fusion have been designed to provide the accurate heading angle information; whereas, in terms of the control module, advanced control technologies such as the linear quadratic Gaussian (LQG) and the model predictive control (MPC) are used to achieve the robust and adaptive tracking performance. However, in the current version of the NGC system, path planning capability has not been fully achieved and integrated. Although in Naeem et al. (2012), the A* algorithm has been applied into Springer, the trajectory was calculated only based upon the distance cost. The absence of more advanced path planning algorithms considering multiple optimisation constraints makes the vehicle only capable of undertaking simple tasks.
In recent days, intelligent systems have attracted many researchers and developers from many branches of engineering and sciences, especially in the field of path planning to bring up innovations in Autonomous motion planning approaches. Their applications vary from data collection, delivery, surveillance to crucial military purposes. Such Autonomous or Unmanned vehicles can be broadly classified into four categories as unmanned underwater vehicles (UUVs), unmanned aerial vehicles (UAVs), unmannedsurface vehicles (USVs) and unmanned ground vehicles (UGVs). Each one has its own application ranging from ocean exploration, remote sensing, and imagery collection to military surveillance. The proposed research work is mainly focused on the intelligent path finding for UGVs. UGV navigation is generally described as a sequence of collision-free movements from starting to the destination position in the provided configuration space. From the study 1 , UGV navigation has three steps:
Unmanned ground vehicles could be navigated by using intelligent control methods.  proposed an neural network for tracking the path of an unmannedvehicle. The input provided to the neural network are velocity, friction coefficient, hope radius, output is velocity difference. For steering control prevention control method is adapted. Different velocity, turning radius and ground surfaces can be adapted by this neural network method. A multi-step neural network was proposed for controlling steering of wheeled mobile robots having complex mathematical model. Neural networks have reduced learning capacity for learning maximum overshoot, overshoot time, steady steering angle and speed. In order to control steering of wheeled mobile robots an GA fuzzy neural network is used. Firstly, a neural network model of mobile robot is developed. The best control parameters are found by using fuzzy neural network and GA method. GA fuzzy neural networks are used for combined speed and direction control of wheeled mobile robots. To achieve this first a controller based on fuzzy neural network is built, then to find best control parameters optimum GA algorithm is used.
In the literature, there are a number of studies that aim to solve several distinct parts of this entire problem space. Murray and Chu  and Ferrandez et al.  proposed a combination of two delivery approaches i.e., the use of tradi- tional delivery trucks to a point near to customers from which a UAV can be launched to deliver parcels. Regarding UAV route/path optimization, Ragi and Chong  analyzed UAV path planning in a dynamic environment via partially observ- able Markov decision process whereas Roberge et al.  studied a comparison of parallel genetic algorithm and par- ticle swarm optimization for real-time UAV path planning. Zhang and Duan  proposed an improved constrained dif- ferential evolution algorithm to generate an optimal feasible route for UAVs as a constrained optimization problem in the three-dimensional environment and Huang et al.  proposed a novel coordinated path planning method using k-degree smoothing for multi-UAVs to reach the targets simultaneously (strong coordination) or with an acceptable time interval (weak coordination). Similarly, Ergezer  explored a path planning in 3D environment for multiple UAVs by introducing an evolutionary operator in Genetic Algorithm and the utilization of mTSP. Yang and Yoo  analyzed the UAV path planning with respect to wireless sen- sor networks (WSN). Several other studies – focused their attention on UAV route/path planning for various objectives, such as target tracking, obstacle avoidance, landmark-basednavigation, cluster-based routing to reduce dependencies on human operators and task assignment.
This paper presents work on the development of a real-time autonomous navigation system for UnmannedSurface Vehicles (USVs). The navigation system being developed is using an embedded hosting platform consisting of navigational data fusion processes together with algorithms used for path planning and collision avoidance when the USV is operating alone or in cooperation. An improved A* path planning algorithm based on rasterized map is developed for single USV operation; whereas the fastmarching square algorithm is implemented for multiple USVs. Both algorithms have been tested using a practical simulation environment. The resulting trajectories are guaranteed to be the shortest collision-free path.
Multi-sensor data fusion (MSDF) for vehicle’s navigation has advanced in recent years; normally a multi-sensor navigation system is hybrid that having both Global Navigation Satellite System (GNSS) and DR system. Most of these integrated systems employ a GPS receiver, several inertial sensors and usually an electronic compass. Some advanced systems would also include sensors like a speed log and perhaps a camera. Caron et al. (2007) proposed particle multi-data sensor fusion algorithms for land vehicle, and concentrated on observing sensors failure and integrated multiple sensors to improve unreliable GPS information. Jared and Gerard (2011) proposed several data fusion algorithms for a GPS receiver and several inertial measurement units (IMUs), which provide good performance when reducing GPS position error. Zhang et al (2005) implement a Kalman Filter to improve the reliability of GPS, IMU and electronic compass measurements. In terms of USV application, Liu et al. (2014) developed a Kalman filter based algorithm to obtain accurate positions, speeds and headings of an USV. However, it should be noted that sensor failure, which is another impact that affects the accuracy of USV’s navigational data, hasn’t been considered in such applications. If one of the sensors fails the consequences could be disastrous since the USV will lose its current situation. An effective method of detecting and disregarding the failed sensor should be considered.
One of the challenges in the development planning of unconventional reservoirs is determining the optimal well spacing. It is essential to understand when and how the well performance has been impacted by surrounding wells. Modeling of well interference in unconventional reservoirs is complicated by the complexity and uncertainties in fracture geometry. In this chapter, we propose a novel and efficient approach based on fastmarchingmethod to identify well interference and quantify the relationship between well spacing and well performance in unconventional reservoirs. The proposed method can directly track the onset of well interference and thus the reservoir can be partitioned accordingly based on the competing drainage volumes amongst the wells. The drainage volume evolution within each subdomain associated with any particular well can be used to recast the 3-D diffusivity equation to a 1-D form which can be solved analytically or numerically for pressure and rate response. As a result, it not only allows us to rigorously compute the well drainage volume as a function of time but also to assess the potential impact of in-fill wells on the performance of existing wells. With these improvement, we then present a new workflow to optimize well spacing in unconventional reservoirs.
Abstract: This paper presents the design, implementation, and testing of a soft landing gear together with a neural network-based control method for replicating avian landing behavior on non-flat surfaces. With full consideration of unmanned aerial vehicles and landing gear requirements, a quadrotor helicopter, comprised of one flying unit and one landing assistance unit, is employed. Considering the touchdown speed and posture, a novel design of a soft mechanism for non-flat surfaces is proposed, in order to absorb the remaining landing impact. The framework of the control strategy is designed based on a derived dynamic model. A neural network-based backstepping controller is applied to achieve the desired trajectory. The simulation and outdoor testing results attest to the effectiveness and reliability of the proposed control method.
 Cao, W., Zhu, L., Han, J., Wang, T., and Du, Y. (2013). High voltage trans- mission line detection for uav based routing inspection. In 2013 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, pages 554–558. IEEE.  Casau, P., Cabecinhas, D., and Silvestre, C. (2011). Autonomous transition flight for a vertical take-off and landing aircraft. In Decision and Control and Eu- ropean Control Conference (CDC-ECC), 2011 50th IEEE Conference on, pages 3974– 3979. IEEE.
Abstract—Path planning of Unmanned Underwater Vehicle (UUV) is of con- siderable significance for the underwater navigation, the objective of the path planning is to find an optimal collision-free and the shortest trajectory from the start to the destination. In this paper, a new improved particle swarm optimization (IPSO) was proposed to process the global path planning in a static underwater environment for UUV. Firstly, the path planning principle for UUV was estab- lished, in which three cost functions, path length, exclusion potential field be- tween the UUV and obstacle, and attraction potential field between UUV and destination, were considered and developed as an optimization objective. Then, on the basis of analysis traditional particle swarm optimization (PSO), the time- varying acceleration coefficients and slowly varying function were employed to improve performance of PSO, time-varying acceleration coefficients was utilized to balance the local optimum and global optimum, and slowly varying function was introduced into the updating formula of PSO to expand search space and maintain particle diversity. Finally, numerical simulations verify that, the pro- posed approach can fulfill path planning problems for UUV successfully.
Although many achievements have been made by research teams, these achievements are mainly concentrated in industrial, transport and other fields, in the agricultural field, especially in the orchard environment, the research of using laser on orchard unmannedvehicle to detect a human obstacle has not been reported. In this research, based on the self-developed orchard unmannedvehicle OUV-2, we established the global coordinate system model, the local coordinate system model and the laser polar coordinate system model, SICK and computer data transmission channel by using TIC/IP communication protocol, the detection distance resolution of obstacle was obtained by calculating the beam diameter and distances between measured points. Combined with the actual working environment of OUV-2, taken a person as an obstacle and conducted the test.
I hereby declare that the work in this thesis is based on my original work except for quotations and citations which have been duly acknowledged. I also declare that it has not been previously or concurrently submitted for any other degree at Universiti Malaysia Pahang or any other institutions.
In 2004, David Buecher made a low cost remotely operated vehicle and his thesis, “Design and Manufacture of a Low Cost Underwater Remote Operated Vehicle(ROV)”, explains how he did it. This is relevant to this project because Buecher’s goals were to make this robot out of commonly found items and for less than $1500. Our goal was to make a smaller and less expensive model than Triton, an existing ROV the RSL uses. Buecher highlighted how he was able to find most of the pieces he needed for the robot at places like Lowes and Home Depot. Anything he could not find inexpensively, he made himself. For example, the tether required to communicate with the ROV that he wanted to purchase was too expensive for his budget so he instead made a neutrally buoyant tether himself (Buecher). The projects are different in that our budget was not as small as Buecher’s. His ROV consisted of motor controllers, an AVR mini board, and a camera. Top-side, he had a computer and Logitech joystick to control the robot via tether, and a VCR to record images from the camera. This thesis helped show how to weigh cost versus quality and helped us maintain our budget.
More recently, the methods underpinned by local descriptors, i.e., Scale Invariant Feature Transform (SIFT), have been employed in image matching widely [7–9]. This approach is regarded as one of the robust feature-based matching methods against the rotation, scaling, and illumination change. Mikolajczyk and Schmid  evaluated the performances of various local descriptors, and found that SIFT outperforms other descriptors for feature matching. However, the computation time will sharply increase when it is applied in images of large size. Moreover, this method is proposed for optical image matching, and issues may arise when it is used in SAR image matching. For instance, the SIFT keypoints extracted from SAR image pairs might be too sparse to calculate transformation parameters and the ratio of false matches will increase greatly.
Unmannedsurface vehicles (USVs) are autonomous marine craft that operate on the surface of a body of water without any personnel onboard. They are analogous to airborne unmanned aerial vehicles (UAVs) and subaquatic unmanned underwater vehicles (UUVs) . USVs have been widely used to conduct scientiﬁc research in the ﬁelds of oceanography  and meteorology  and have their applications in the oil and gas industry also. Within the Defence sector, USVs are currently being developed for several roles including anti-submarine warfare and minesweeping. One such USV is Halcyon which is currently being developed by Thales UK and ASV Global for autonomous mine clearing missions. The simulation model presented in this paper has been developed to aid in the development, testing and validation of Halcyon’s autonomy management system. Using simulation for this purpose reduces the need to conduct time-consuming and expensive sea-trials and allows for greater ﬂexibility over the environmental conditions in which the boat must operate. This ﬂexibility oﬀers the additional advantage of being able to test and evaluate several guidance, navigation and control (GNC) systems using the same “random” wave environment. To aid in this, the simulator incorporates a novel sea-surface wave environment model which is an integration of
In short, when using computer in learning, student tend to learn fast and better memorization. Besides that, computer has positively affected student’s attitudes toward learning and school. Students result does not guarantee with the computer and computers-related technology but also included many other factors that play important roles such as instructional design and software complex. Besides that, an intelligent tutoring system with the add-on of advanced planning and natural language will become the focus of the new generation (Fouts, 2000).
diffusivity condition, all the differences on the w ( ) plot is attributed to fracture geometry. During the formation linear flow period for infinite conductivity scenario, the drainage volume increases at a rate which is proportional to the fracture surface area. Therefore, we see constant value on the w ( ) plot at the early time and the cases with 60 clusters show about 50% larger value for the cases with 40 clusters (note that the fracture half-lengths are the same). When the fractures inference with each other, the drainage volume increase slower and the w ( ) starts to decline. The case 3 with cluster spacing 75 ft shows about 50% larger value compared to cases with cluster spacing 50 ft at the same strength of fracture interference. The second section is due to that the pressure continues propagating in the reservoir as compound linear flow. The results are purely from synthetic simulations; for field cases, where one horizontal well is drilled near another, we don’t expect the second section. What’s more, due to finite conductivity effect (eg. proppant degradation) or partial penetration effect (eg. proppant sit lower part of the fracture), we don’t expect strong flat feature at early part followed by sharp drop, instead the ( ) w
Considering both the segmentation quality and the com- putational cost, in this paper, we propose an e ﬃ cient ap- proach to ROI extraction. Diﬀerent from the other ap- proaches, neither statistics are needed to be computed con- tinuously nor complex numerical implementation is in- volved. The deforming curve is modeled as monotonically marching front under a new positive speed field, where a new region speed function is derived by minimizing the ROI energy. Integrating with the region information, the modi- fied speed function has large propagation range and could even drive the front propagating in low-contrast and nar- row thin areas. To further improve the segmentation re- sults, multi-initial scheme is adopted  and the multi- initial fastmarching algorithm is developed, which permits the user to plant several seed curves as the initial front and evolves them simultaneously. All the seed curves are treated as one complex front driven by the same evolution equa- tion. Selective planting seed curves can avoid the monoton- ically marching front leaking out of the weak boundary too early to arrive at the desired boundary and it can also reduce the computational cost. Our approach is similar to that of Vilari˜ no’s cellular neural networks (CNN) approach to im- age segmentation [18, 19]. Both approaches evolve pixel by pixel from their initial shapes and locations until delimit- ing the objects of interest, and the curve evolution is guided by local information from the image under consideration, which can oﬀer a high flexible and eﬃcient parallel process- ing.
GPS is used for localization of intelligent vehicles. GPS consists of 24 satellites which send signals to estimate position. One satellite needs to be received for each dimension of the user‟s position that needs to be calculated. This suggests three satellites are necessary for position estimate for general user (for the x, y, and z dimensions of the receiver's position) however, the user rarely knows the exact time which they are receiving at, hence four satellite pseudo-ranges are required to calculate these four unknowns. The satellite data is monitored and is controlled by the GPS ground segment - stations positioned globally to ensure the correct operation of the system. The user segment is the GPS user and the GPS reception equipment. These have advanced considerably in recent years to allow faster and more accurate processing of received data. They typically contain pre-amplification, an analogue to digital converter and DSP processors etc. .
There are two kinds of mathematical models for ship motion. One is the Abkowitz model [10,11], which is also called the global model. It considers the hull, propeller and rudder as an integrated whole, and expands the hydrodynamic force acting on the hull into the Taylor series of each motion variable. The other is the so-called MMG model [12–14], which is also known as the separable model. It was proposed by the Japanese Mathematical Modeling Group (MMG) based on the Abkowitz model. The MMG model decomposes hydrodynamic forces into three parts of the hull, propeller and rudder, and takes the interaction between them into account. The Abkowitz model is mathematically more complete and rigorous, and therefore widely used in the community. Based on the Abkowitz model, Fossen et al. [15,16] put forward a method of modeling and control for marine craft, including ships, high-speed craft, semi-submersibles and floating rigs. They conducted a detailed analysis of the kinematics and kinetics performance of the marine craft. J. Menoyo Larrazabal et al.  formulated a coupled surge-sway-yaw model to guide the USV maneuvering control and dealt with the rudder control of a USV system. Based upon a simplified 3 degrees of freedom (DOFs) model and an actuation model, Petr Švec et al.  presented a trajectory planning and tracking approach to follow a di ff erentially constrained target vehicle operating in an obstacle field. Simulation and experimental studies were conducted to demonstrate the e ff ectiveness of the developed method.