Design of an Underwater Glider Platform for Shallow-Water Applications

Transcription

1 Design of an Underwater Glider Platform for Shallow-Water Applications Nur Afande Ali Hussain 1, Ting Ming Chung 2, Mohd Rizal Arshad 1, Rosmiwati Mohd-Mokhtar 1, Mohd Zulkifly Abdullah 2 1 Underwater Robotics Research Group, School of Electrical and Electronic Engineering Universiti Sains Malaysia, Engineering Campus Nibong Tebal, Seberang Perai Selatan, Pulau Pinang, Malaysia 2 School of Mechanical Engineering Universiti Sains Malaysia, Engineering Campus Nibong Tebal, Seberang Perai Selatan, Pulau Pinang, Malaysia Abstract Underwater gliders are type of autonomous underwater vehicle that glide by controlling their buoyancy and attitude using internal actuators. By changing the vehicle s buoyancy intermittently, vertical motion can be achieved. Characteristic of glider motions include upward and downward in a saw tooth pattern, turning and gliding in a vertical spiral motion glides without using thrusters or propellers. This paper presents the development of the USM underwater glider as the first prototype for shallow water applications. The prototype development involves vehicle concept design using Solidworks, vehicle simulations by Computational Fluid Dynamics (CFD) and MATLAB Simulink as stage of the design process. Once the prototype fabrication and system integration are completed, it will be tested for vehicle s modeling and controller development using the system identification approach and will be compared with the proven glider s control model. Keywords Underwater Glider; Computational Fluid Dynamics; 1.0 Introduction An underwater glider glides by controlling their buoyancy using internal tanks and pumps. Existing gliders have fixed external wings and tails and control their attitude by moving internal masses and using external control surfaces such as rudder. Gliders travel from place to place by concatenating a series of upward and downward glides. Gliding flight is buoyancy driven, and does not use thrusters system. Thus, glider must change depth to glide. They glide downwards and upwards in the ocean by controlling their buoyancy to make themselves negatively and positively buoyant. Gliders may also hold their position by gliding against current. This will make them neutrally buoyant and drift with the current or rest on the bottom. Through their use of buoyancy propulsion system and low power design, gliders are capable of long-range and high-endurance deployments. The driving force of the glider depends on hydrodynamic forces such as drag and lift, which is a function of the advance speed and the angle of attack. The gliding nature of about one half knot while travelling makes it relatively slow in comparison with other

2 Autonomous Underwater Vehicles (AUV) that used the propellers. However, the gliders (without the propellers) have longer ranges, reduce the underwater acoustics signals and are more difficult to detect. Gliders can operate both in littoral (coastal) and very deep ocean environments. Depends on the control system, the operations could involve gliders to maneuvering in the interest area for long periods of time. Using the suitable control algorithm, it can be programmed to do transiting undetected between areas or waiting on the bottom for several periods of time before beginning a new mission command. The emerging technology of gliders has overcome many difficulties that lie on the conventional method of ocean sampling using the ships and drifters. For instance, ships can be too expensive to be used and limited in number and availability while drifter cannot choose their path through ocean. Satellite technology in navigation and communication can be utilized on gliders application and changing the way we see the ocean environment. These platforms can provide a different view of interior ocean with much higher spatial from the command center situated from far away. Underwater glider design and development approach can be different from one another due to user requirements, amount of funding and man power involved. It also relies on whether some user needs to utilize high sophisticated sensory system and amount of prototype endurance in term of duration of mission and maximum gliding depth. Some of the design examples can be seen as in (Arima, Ichihashi and Ikebuchi, 2008; Alvarez et.al, 2009, Graver, 2005; Saringer, 2009, Ross, 2006). Early development of prototype design using simulation can be useful to investigate the vehicle behavior. CFD analysis can be useful for hydrodynamic forces approximation and to study the prototype performance in advanced (Seo, Gyungnum and Choi, 2008; Ishak, 2006). We can study the prototype shape to find the suitable design and formalized the internal component needed as a payload. CFD simulation result can be verified based on the towing tank experiment using the force sensor. It measures the real hydrodynamic forces and compared with the simulation results and shows high in correlation between them (Arima, Ichihashi and Ikebuchi, 2008). The simulations result can be used as the tool for design and development in cost effective way for underwater glider application. Challenge in control system for underwater glider is the underwater environment, in which it can be classified as unstructured and unknown. Suitable control methodology and strategies can be chosen to overcome these unknown parameters. One of the control strategies is to divide the control architecture into three specific layers such as mission planning layer, task planning layer, and the behavior layer (Yu, et al., 2006). This type of controller integrates the human s intelligence into the control system and can assure the preciseness in planning level. There is also some dedicated controller design for sharp turning strategies using Linear Quadratic Regulator (LQR) for underwater glider during reflection near the sea bottom and collision prevention (Wang, Zhang and Wang, 2009). Robust PID controller worked effectively during diving control phase in autonomous underwater vehicle combining glider capabilities (Alvarez et al., 2009). Apart from this, there are also research undertaken on model-based feedback control for gliders as for example in Leonard and Graver (Leonard and Graver, 2001) and the study on stability and control of gliding vehicle as being discussed by Bhatta (Bhatta, 2006). With careful design, buoyancy-driven gliders are quiet and used little power (Graver, 2005; Leonard and Graver, 2001). Underwater gliders can be utilized in remote sensing for physical, chemical and biological oceanography. Other possible applications include used as communications gateways or navigation aids and military application such as tactical oceanography and maritime reconnaissance. Attractive glider characteristics include of autonomous operation, high endurance and range, and low cost. Gliders are well suited to extended missions such as collecting sensor data while move along a chosen path or maintaining position. Gliders can operate autonomously either as individual task or in a group and may adaptively adjust their missions according to remote instruction or according to sensor information. Based on the remarkable issues that glider can overcome, together with the positive advantages and wider application area that can be utilized, therefore, developing another new platform for glider system will become the

3 interest of this paper. In summary, this paper goes as follow. Section 2 describes the system design of the USM underwater glider prototype. In section 3, system implementation of the prototype design, CFD simulation analysis and simulation equation of motions are elaborated. Section 4 explains the discussion of the findings. Finally, conclusions are given in section System Design of the Underwater Glider The USM underwater glider is under development in order to be the first prototype platform to investigate the buoyancy driven mechanism and vehicle dynamic motion. The platform is designed to be operated on shallow sea water not more than 30 meters depth (Ali Hussain, Arshad and Mohd-Mokhtar, 2009). To complete the design, fundamental knowledge on flight dynamic is adopted as it shares some common features in terms of the system dynamics (Etkin, 1959). The careful and deep understandings on fluid mechanics are also important as to make sure the developed platform is robust and reliable with the real practical situation (Douglas, Gasiorek and Swaffield, 2001; Yunus and Cimbala, 2006). Upon completion of the prototype, it will be tested for it workability in giving the forward motion. The shape of the underwater glider will be in circular cylindrical form based on the proven design concept. This is due to underwater hydrodynamic pressure effect and the capability of this shape to resist this effect efficiently. Another advantage is a good hotel load for the hardware such as electronic components. Extra payload can be achieved by designing the shape hull longer or in modular shape. This will give better accessibility on different subsystem too. As for the same volume, a circular cylinder provides with better hydrodynamic as compared to spherical form. For the docking purposes, circular cylindrical shape is easy to implement than others. Figure 1 and Figure 2 display the hull shape design and the internal frame design respectively. The overall length of the vehicle is 1.3 meters with a diameter of 0.17 meter. In this underwater glider design, National Advisory Committee for Aeronautics (NACA) 0012 airfoils as a wing geometrical construction is selected. The first two digits indicate the symmetrical of the airfoil and the last two digits indicates the maximum thickness percentage compare to it chord size. NACA airfoil series are the guidelines being used to construct the wing airfoil coordinates. Although these concepts of constructions are restricted for airplanes design, it can be adopted to the underwater environment due to gliding concept as propulsion mechanism for the vehicle. These wings design are crucial to give the vehicle sufficient Lift-Drag ratio as a conversion from vertical velocity to forward motion. The total wing span for this vehicle is 1 meter without sweep design. Table 1 gives the general specification for the USM underwater glider. Table 1- Specification of USM Underwater Glider Dimension Operation Depth Operation Time Main Power Sensors 0.17 m (Diameter) 1.3 m (Length) 1.0 m (Wing Span) 30 meters maximum More than 2 hours Lithium-Ion Eco-Sounder Transducer IMU 5 Degrees of Freedom Gyrocompass Depth Sensor Distance Sensor

4 Figure 1- Hull shape Design Figure 2 Internal Frame Design The pressure hull is made of high pressure PVC for easy manufacturing process. Since there is no propellers system in this platform, the buoyancy system and movable mass mechanism behavior will be monitored by the data logger system for workability of the mechanism during underwater testing stage. The maximum weight of USM underwater glider will not exceed 30kg including all payloads, future sensors and instrumentation equipment. The design also includes a future expansion of multi sensor such as two acoustic altimeters and Conductivity, Temperature and Depth (CTD) module for auto-heading, auto-depth and auto-altitude capability, as well as for survey mission. Mission control system such as diving control and surface navigation will be developed in order to do the heading, depth and altitude mode. It shall be portable to handle by one or two personnel for ease of transportation and due to Launch and Recovery (LAR) aspect. With this portability system, multiple gliders can be launched in a mission without needed any special mechanical equipment. By having this, the payload of all equipments can be reduced significantly (Jun et al., 2009). 3.0 System Implementation of the Underwater Glider 3.1 Electrical and Sensors System PIC programmable chip will be used as the main controller circuit as it consumes less energy for endurance of the vehicle. For heading and orientations of the vehicle, Inertial Measurement Unit (IMU) of 5 Degree of Freedoms (DOF) provides the sensing of surge, heave, sway, rolling and pitching. This will give some important decision

5 making for control design and system identification processes. By using a CruzPro active acoustic transducer (see Figure 3) and KELLER depth sensor, the vehicle can be programmed to execute the auto-depth and auto-altitude routine. Due to the slow horizontal motion of the glider, these capabilities will allow the vehicle to perform underwater profiling over a quite large number of areas. For each sensor and actuator, PC-104 will be fitted to the vehicle for future data logging system. The main power to the vehicle is 9 AH Lithium Ion battery that able to support up to 3 hours of operation. 3.2 Buoyancy engine Figure 3- CruzPro Acoustic Transducer The buoyancy engine system of the underwater glider is based on the ballasting tank which has a piston to push out and pull in the water inside the vehicle. The piston tank is driven by 12V DC brush motor and capable of injecting maximum of 500ml of water in 30 meter depth. The motor can be changed to high torque DC motor in order to improve vehicle capabilities to go deeper and to overcome the underwater pressure. The ballasting mechanism is monitored by using micro switch to estimate the ejecting volume of water. The micro switch is able to estimate the internal piston location for detail dynamic analysis as this system provides the desired depth and altitude control. This DC motor is driven by the 10A driver circuit, being interfaced with the micro processor chip, which act as a controller for the system. Figure 4 and Figure 5 show the enhanced 10A motor driver and the piston tank respectively. Figure 4- Enhanced 10A Motor Driver

6 3.3 Movable Mass System Figure 5- Piston Tank The movable mass platform consists of stepper motor control mechanism for precise controlling position of the platform. The platform will be integrated with the Lithium-Ion battery for movable mass to save the payload space. Stepper motor is integrated with the infrared distance sensor in order to know the exact position of the movable platform. Again the data can be analyzed to gain the dynamic behavior of the vehicle. 3.4 Vehicle Design Based on the calculation of sea water density of 1027kg/m 3, the minimum vehicle weight is around 23kg in order to make it slightly in negative buoyancy. It is of important also to consider all the volume of the vehicle inclusive of the internal component mass and the vehicle size in order to estimate overall shape of the vehicle. The center of gravity (CG) will give some estimation on proper place to install the movable mass system into the underwater glider. Calculations are done by using Datcom software. CG will be estimated according to the current components arrangement. In future, all the add-on components will be arranged in reference to this CG. Table 2 shows the calculation of CG based on the location of sensors and hardware components. In this case, Y axis is the length of the vehicle. CG in X direction, CG in Z direction, In these calculation, the CG are located at the (821.6mm, 0, -0.22mm) coordinate point.

7 Table 2- Example of CG Calculation COMPONENT WEIGHT, M (KG) Distance, X (mm) M x X Distance, Z (mm) M x Z PISTON TANK VOLUME VOLUME VOLUME BATTERY BATTERY EACH BATTERY PC OR 0.5 EACH CFD Analysis In performance analysis of underwater glider, estimation of hydrodynamic forces is important. As underwater glider is forced to move through fluid, drag and lift forces that generated will have significant effect on the motion and performance of the glider itself. The drag and lift forces depend on the density of fluid, velocity of fluid, size, shape and orientation of vehicles body. Drag force coefficient, C D and lift force coefficient, C L are dimensionless number which represent drag and lift characteristics of the vehicle. Drag force coefficient, C D =F D / ( Lift force coefficient, C L = F L / ( (1) (2) Here, A refers to the platform area of main wings. For small angle of attack (positive and negative), Fluent is used to run the simulation. The resultant drag and lift force is transformed into drag and lift coefficient according to the equations (1) and (2) mentioned above. For glider modeling, it cuts into half at the symmetry plane. This is used to shorten the simulation time and reduced memory usage. Figure 6 and Figure 7 demonstrate some results of lift and drag coefficient for different angle of attack. For underwater glider at zero angle of attack, it shows that the total pressure distribution is quite symmetrical especially at the body of the vehicle. Waves occurred at the back region of the glider as fluid pass through glider body. Figure 8 shows some contour on total pressure for underwater glider at zero angle of attack.

8 LIFT/DRAG RATIO Int. J. Intelligent Defence Support Systems, Vol. 3, No. 3, 2010 Figure 6- Lift And Drag Coefficient Vs Angle Of Attack LIFT/DRAG RATIO VS ANGLE OF ATTACK LIFT/DRAG RATIO -4 ANGLE OF ATTACK Figure 7- Lift/Drag Ratio Vs Angle Of Attack Of Glider Total Pressure On Symmetry Plane,Zero Angle Of Attack Contour Of Total Pressure, Zero Angle Of Attack Figure 8- CFD Analysis

9 Figure 9.1- Bottom View Of Total Pressure Distribution, Upward 2 Degrees Figure 9.2- Bottom View Of Total Pressure Distribution, Upward 10 Degrees Figure 9.3- Top View Of Total Pressure Distribution, Upward 2 Degrees Figure 9.4- Top View Of Total Pressure Distribution, Upward 10 Degrees

10 Figure 9.5- Bottom View Of Total Pressure Distribution, Downward 2 Degrees Figure 9.6- Bottom View Of Total Pressure Distribution, Downward 10 Degrees Figure 9.7- Top View Of Total Pressure Distribution, Downward 2 Degrees Figure 9.8- Top View Of Total Pressure Distribution, Downward 10 Degrees Figures show a total pressure distribution of underwater glider for upward and downward Angle of Attack (AOA). As angle of attack increases positively from two to ten degree, total pressure on bottom surface increases to nearly factor of two especially at the location of fixed wing and tails. Nose of glider and leading edge of fixed wings has shown the highest of total pressure distribution. For upper surface of glider, changes of total pressure are less. However, region with low pressure has increased (blue region on main wings area) accordingly with angle of attack. This is the area where main wings joint with glider body. For diving condition, pressure force acting on upper surface is higher as lift force is acting downward. As angle of attack increases negatively, total pressure acting on upper surface is also increased. For velocity flow profile, it shows a shock wave appears at the back of the glider as the fluid passing through. For upward AOA, as angle of attack increases, the shock wave behind the glider expands as well, and it spreads

11 towards upper surface of the glider. For downward angle of attack, shock wave expands in different direction which spreads to lower surface of the glider. Pressure force acting on upper surface of the glider is higher than lower surface. Nose of glider also shows a high resistance against the fluid flow where pressure distribution is very much higher as compared to other region. Upward 2 Degrees Upward 10 Degrees Downward 2 Degrees Downward 10 Degrees Figure 10- Velocity Profile 3.6 Simulation of Equation of Motions Simulation of equation of motions in the vertical plane was done using the MATLAB Simulink in order to investigate the buoyancy driven mechanism in advanced (Graver, 2005; Kan et al., 2008; Gardner, 2001). The performance of the vehicle is analyzed by simulation, and this, can be very useful for glider design and behavior understanding. The simulation result will be used for validation purposes, while comparing the real experimental data with the proven underwater glider model (Leonard and Graver, 2001; Bhatta, 2006).

12 Figure 11- Forward Velocity (m/s) Figure 14- Lift Figure 12- Pitching Angle (radians) Figure 15- Drag Figure 13- Angle of Attack (radians) Figure 16- Depth (meter) In this result, we have selected the gliding downward motion with zero initial condition. The controlling conditions are: Forward Velocity=0.3m/s, Pitch Angle=-0.5 radians and Net Buoyancy =25 grams. During the 500

13 seconds of simulation, the vehicle is gliding downward towards the 40 meters depth. Based on these results, we are able to compare with the CFD findings, and this will become the additional data information for this research. The simulation results have been justified with similar test field conducted by (Graver, 2001). Figure 11 to Figure 16 show some of the glider dynamics based on the MATLAB Simulink design investigation. 4.0 Discussion The USM underwater glider is the first glider prototype that has been developed at the Underwater Robotic Research Group (URRG) USM. The vehicle can be occupied with sensors payload depending on the purpose and applications. In this prototype, we obtained the model of the glider s dynamics by using the system identification approach. Using the low cost sensors it is sufficient to know the vehicle dynamics based on the manipulated control inputs. Ecosounder transducer is used as the alternative for low cost altitude data collector as compared to the proven expensive altitude transducer. The simulations results which represent the vehicle dynamics and behaviors have provided with important information for the understanding of the vehicle behavior in general. In CFD analysis, the findings are based on the shape of the vehicle design whereas in MATLAB simulations, it is based on the control inputs configuration. By combining all these results, we may get the complete picture of the system design overall. All the gathered information will be very useful for future underwater prototype development. 5.0 Conclusion Based on the results and findings, we have gathered enough information for developing our first prototype of glider system. The CFD simulation results show the gliding motion in different AOA. It also demonstrates some possible gliding angle as to investigate the vehicle movement during gliding. The total pressure distribution to the vehicle s body while gliding is also observed. This information will be used in manufacturing and fabrication process for higher stresses area. The simulation of motion in the vertical plane shows some vehicle characteristic of the glider motion based on the ballast rate and movable mass displacement. These inputs will manipulate the vehicle characteristic in terms of depth, gliding angle, forward velocity etc. Based on the findings, the system identification approach is used to obtain the vehicle s control system model. Later, the outcome based on the system identification approach will be compared to the simulation model to give information about the input and output parameters. The control system hardware can be implemented to the prototype so it can be workable and reach its optimum criteria according to the prototype design aspect. Acknowledgement The authors would like to thank the National Oceanographic Directorate (NOD) of the Ministry of Science and Innovation (MOSTI), NOD-USM and USM-RU-PRGS under Intelligence Control & Modeling of Underwater Glider for the research grant awarded. References Arima, M., Ichihashi, N., and Ikebuchi, T., (2008). Motion Characteristics of an Underwater Glider with Independently Controllable Main Wings. OCEANS 08 MTS/IEEE Kobe-Techno-Ocean 08- Voyage toward the Future, pp.1-7.

14 Ali Hussain, N., A., Arshad, M., R. and Mohd-Mokhtar, R., (2009), Development of an Underwater Glider Platform. Proc. of the Electrical and Electronic Postgraduate Colloquium EEPC2009, 2 nd November 2009, Penang, Malaysia. Alvarez, A., Caffaz, A., Caiti A., Casalino, G., Gualdesi, L., Turetta, A., Viviani, R. (2009). Fòlaga: A low-cost autonomous underwater vehicles combining glider and AUV capabilities. Ocean Engineering, 36 (1), pp Bhatta, P., (2006). Nonlinear Stability and Control of Gliding Vehicles. PhD thesis. Dept. of Mech.& Aerospace Eng., Princeton University. Douglas, John F., Gasiorek, Janusz M., Swaffield, John A. (2001) Fluid Mechanics, 4th Ed., London: Addison- Wesley. Etkin, B.(1959) Dynamic of Flight. New York: John Wiley and Sons. Gardner, J. F. (2001) Simulation of Machines, Using MATLAB and Simulink. Canada: Wadsworth Group. Graver, J. G. (2005). Underwater Glider: Dynamic, Control and Design. PhD thesis, Dept. of Mech.& Aerospace Eng. Princeton Univ. Ishak, I.S. (2006), Computational Fluid Dynamics Simulation and Wind Tunnel Testing on Microlight Model, 1st Regional Conference on Vehicle Engineering & Technology, 3-5 July 2006, Kuala Lumpur, Malaysia. Jun, B. H., Park, J. Y., Lee, F. Y., Lee, P. M., Lee, C. M., Kim, K., Lim, Y. K., and Oh, J. H., (2009). Development of the AUV ISiMI and free running test in an Ocean Engineering Basin. Ocean Engineering, 36 (1), pp Kan, L., Zhang, Y., Fan, H., Yang, W., and Chen, Z., (2008). MATLAB-Based Simulation of Buoyancy-Driven Underwater Glider Motion. J. Ocean Univ. Chin.(Oceanic and Coastal Sea Research).Vol 7, No.1, pp Leonard, N. E. and Graver, J. G., (2001). Model-Based Feedback Control of Autonomous Underwater Gliders. IEEE Journal of Ocean Engineering, Vol. 26, No. 4, pp Ross, C. T. F. (2006), A Conceptual Design of an Underwater Vehicle. Ocean Engineering Volume 33, Issue 16, pp Saringer, A. (2009), Hypersonic Glider For A Wave-Rider Vehicle, Old Dominon University. Obtain through Internet: [1,10,2009]. Seo, D.C., Gyungnam Jo, and Choi, H.S. (2008), Pitching Control Simulation of an Underwater Glider Using CFD Analysis. OCEANS (2008)-MTS/IEEE Kobe Techno-Ocean, pp Wang, Y., Zhang, H., and Wang, S. (2009), Trajectory Control Strategies for the Underwater Glider. International Conference on Measuring Technology and Mechatronics Automation, ICMTMA 09 Volume 1, April 2009, pp Yu, Z., Jiaping, T., Donghai, S., and Shijie, W. (2006). Research on the hierarchical Supervisory Control of Underwater Glider. IEEE Int. Conf. on Intelligent Robots and Systems, pp Yunus, A.C and Cimbala, J.M. (2006), Fluid mechanics fundamentals and applications, New York: McGrawHill.