Idp Report Compile.docx

MINI AIR VEHICLE WITH FLAPPING MECHANISM INTEGRATED DESIGN PROJECT DEPARTMENT OF MECHANICAL ENGINEERING UNIVERSITY OF MALAYA SEMESTER 1 2018/2019 MUHAMMAD ARIG AIMAN BIN MAKMON

KIG160074

ABDUL MUHAIMIN BIN MAZELI

KIG160002

MUHAMMAD HAZIQ BIN AHMAD SUHAIRI

KIG160086

MUHAMMAD SIDDIQ MUZAMMIL BIN SAMSUL KAMAL

KIG160099

CONTENT Abstract Acknowledgement Management Chart Chapter 1

xx xx xx

A. B. C. D. E.

Introduction Project background Literature review Problem statement Objectives Scope and limitations of the projects

xx xx xx xx xx

A. B. C. D.

Feasibility Study/ Market Survey Market analysis Customer analysis Product benchmarking Technology benchmarking

xx xx xx xx

A. B. C. D.

Concept generation Defining design problems Solution exploration Creative thinking method Concept sketch

xx xx xx xx

Chapter 2

Chapter 3

Chapter 4

Concept Selection A. Concept screening B. Concept scoring C. Concept analysis Conclusion References

xx xx xx xx xx

ABSTRACT

ACKNOWLEDGEMENT I would like to express the deepest appreciation to my Faculty of Engineering, University of Malaya, Dean, Prof. Ir. Dr. Abdul Aziz bin Abdul Raman, who have allowed us to commence this kind of research with sufficient amount of time and money. Without this kind of help a lot of problem will occur and the research will be not perfect as it is now.

I would like to thanks to our supervisor for this project, Dr. PooBalan Ganesan and his sub coordinator, who have showed their support and enthusiasm to develop a new design of Flapping Mini air vehicle. They continually and convincingly conveyed a spirit of adventure in regard to research. Without their guidance and persistent help this research would not have been possible.

In addition, thank you to Aerospace Engineering Lab assistant, Mr. Harron Wardak, whose continually give us the spirit to fight back from comment and issues raised during our weekly meeting. Without your comment and advised our project will not go far as it is.

MANAGEMENT CHART

Dr. PooBalan Ganesan

Sub Coordinator

Team Member

Dr. Azlin Dr. yap Dr. Raja Dr. Andri

Arif Aiman Makmon Haziq Ahmad Suhairi Siddiq Muzammil Samsul Kamal Muhaimin Mazeli

CHAPTER 1 INTRODUCTION

INTRODUCTION Micro Aerial Vehicles (MAVs) are a group of Unmanned Aerial Vehicles (UAVs) that are significantly smaller in the overall dimensions and weight, with a size of approximately six inches. There has been much interest in MAVs especially for applications where maneuverability in confined spaces is necessary, i.e. internal inspection of pipes, exploration around rubble in collapsed buildings and surveillance of indoor environments. Other missions of interest for MAVs include detection, communications and placement of unattended sensors. Due to the availability of very small sensors, detection missions include the sensing of biological agents, chemical compounds and nuclear materials, i.e. radioactivity. Because these vehicles may fly at relatively low altitudes, i.e. less than 300 ft where buildings, trees, hills, etc. may be present, maneuverability is an important factor to avoid collision. Furthermore, since these vehicles have essentially small flying wings, there is a need to develop efficient low aspect-ratio wings which are not overly sensitive to wind shear, gusts, and roughness produced by precipitation. In the mid 90’s, the Defense Advanced Research Projects Agency (DARPA) started a research program that led to the development of the first serious generation of MAVs. These vehicles were airplane-like models with fixed wings, propeller driven and could carry a battery, cameras, R/C transmitters and micro servos for flight control, and could fly for about 20 minutes. They typically weighed around 50g, extended less than 6 inch in all dimensions and were almost exclusively characterized by a rigid wing design with discrete rudders and flaps actuated by micro servos via a rod or wire system. There are three general trends of MAV under investigation: (a) airplane-like models with fixed wings, (b) helicopter-like models with rotating wings, and (c) bird or insect-like models with flapping wings. For that developed a MAV named BatRider, with a flapping mechanism that adapted from bat. Bat can fly with high maneuverability with the wingspan mechanism. In addition, with the flapping mechanism of the wing, a flying platform will be more resistant to shocks so it can fly more stable. The platform to be designed is an unmanned aircraft that controlled by remote control.

PROJECT BACKGROUND Our team is far from the first group to take on this challenge. There have been numerous other attempts to replicate flapping bird and insect flight from a mechanical device. Notable projects that we examined were the the Mentor at University of Toronto, the Delfly at Delft University of technology and also the RoboBee at Harvard University. The University of Toronto’s Mentor robot and Delft University of Technology’s Delfly both use parts modeled on natural flapping wing animals as well as integrating parts similar to conventional aircraft. Both Delfly and Mentor are biplane MAVs utilizing four wings in what is known as a clap and fling method for obtaining lift and thrust. Meanwhile, RoboBee project from Harvard University have made great step in developing robot capable of flight and hovering that is similar weight and size to a common insect. 1.1.1

University of Technology Deft, Delfly

Delfly has existed in three finest form. The first form of this MAV is Delfly I that has capabilities forward flight as well as slow near hovering-flight. The Delfly has dimension of wingspan of 50cm and weight of 21g. Same with all prototypes, Delfly I suffered a lot number of problem relating to control, stability and also reliability. Its invert V-tail allowed for stable forward flight but it affected the difficulty to controlling altitude of this MAV.

Figure 1: DelFly I (right side), DelFly II (left side) and DelFly Micro Delfly I’s drive train was designed so that as the motor turned, a Z shaped crank would rotate causing the wings to move up and down. This method was difficult to sync and resulted in the craft experiencing

slight rolling during flight. In addition, the motor did not possess a high efficiency which, along with it not being brushless, caused it to build up heat during flight. This decreased reliability caused for frequent motor replacements.

Figure : Delfly I Drivetrain The project was later redesigned with many of these flaws corrected into the DelFly II. This model was lighter and smaller, weighing only 16g with a wingspan of 28cm. The drive train was completely redesigned to eliminate the rolling motion present in DelFly I. In addition , the revamped design included custom-made high-performance components, such as a custom brushless motor capable of much higher efficiencies and a custom microcontroller for experiments with autonomous flight. The tail was redesigned to use a classic cruciform tail seen on most model aircrafts to allow for pitching and yawing motions which the vehicle was previously incapable of. This vehicle was capable of both indoor and outdoor flight, as well as maintaining the stability required to have a camera as payload. This design has been the subject of numerous research projects on optimizing the flapping motion and increasing lift as well as various controls and autonomous flight experiments.

1.1.2

University of Toronto Mentor

Figure: Toronto Mentor Project Mentor at the University of Toronto was among the first radio controlled MAV’s to achieve hover using 4 wings in a clap and fling motion, as shown above in figure. Mentor proved the viability of the clap and fling method and the possibility that such a device could be stable while under remote control without any sort of autopilot. Mentor was designed to have a similar wingspan at 26cm but a much higher weight than our project calls for. Mentor had two configurations: one with an internal combustion engine (ICE) and another on battery power. The ICE method weighed 580g – a fourth of that being the fuel and motor. The battery powered method could not make use of relatively recently developed Lithium-Ion batteries but instead used much more inefficient Nickel Cadmium. This resulted in a weight of 440g – over half of which was allocated to the motor and batteries. Mentor was never intended to emulate any actual bird or insect. It was proof the clap and fling method could be used to achieve stable hover and lift even 3 with a relatively heavy aircraft. This inspired the WPFly design as it pointed the group in the direction of the fling and clap research.

1.1.3

Harvard University Robobee

Figure: Harvard Robobee Project

Robobee project at Harvard University was recently developed a tiny robot that very closely similar to flapping insects. The wings have three degree of freedom by making use of piezoelectric actuators. The robot does not have a tail and makes its attitude adjustments through manipulating its wings. This is the same as how an actual insect flies and controls itself. The robot suffers from scaling issues and the device requires external power, for no commonly available power source is small enough to fit onboard. The lack of a tail and any sort of control sensor also results in unstable flight; most tests only run for a few seconds until the aircraft loses control. RoboBee is a marvel of engineering, using novel miniaturization and manufacturing techniques; however, with a wingspan of just one inch, its scale is smaller than the scope of this project and was not heavily considered in our design.

OBJECTIVES The objective of this project is to develop light weight of a model Micro Aerial Vehicle (MAV). This MAV should have the best control and maneuverability compare to other flapping MAV encountered before.

PROBLEM STATEMENT MAV are not only scaled down versions of larger aircrafts “they are affordable, fully functional, militarily capable, small flight vehicles in a class of their own” [6]. With their reduced size, they have to keep all the features of larger aircraft in a small volume, which increase the complexity and challenges. However, in the last few years, the miniaturization progress of AVS has practically stopped [4] (See Figure 1). This mainly happened since there are several problems. There are both physical and technological challenges that slow down further miniaturization [4].

Figure 1: Air Vehicle System (AVS) development reduction; data from 1998 to 2002 [4] and 2008 to 2009

The first problem that appears is related aerodynamics related to the low Reynolds number for AVS. This dimensionless number reflects the ratio between the inertial forces and the viscous forces and is defined as 𝑅𝑒𝑦𝑛𝑜𝑙𝑑𝑠 𝑛𝑢𝑚𝑏𝑒𝑟 =

𝑓𝑙𝑢𝑖𝑑 𝑑𝑒𝑛𝑠𝑖𝑡𝑦 × 𝑠𝑝𝑒𝑒𝑑 × 𝑠𝑖𝑧𝑒 𝑣𝑖𝑠𝑐𝑜𝑠𝑖𝑡𝑦

For AVS, both speed and size are several orders of magnitude smaller than for large aircrafts. This gives Reynolds number, less than one-hundred thousand, which is less than one-tenth of what is common for a full-size aircraft (Figure 2). Flight in this aerodynamic domain is more difficult. Since other physical laws are governing in this domain, a lot of efforts have been made to understand ultralow Reynolds number flight, studying the flight of insects whose size is tiny.

Figure 2: Reynolds number for aerial vehicles, adapted from [].

Although aerodynamics at low Reynolds numbers are not clearly understood yet [5], it is well know that for Reynolds numbers under 100,000, the aerodynamics efficiency (defined as lift-to-drag ratio) rapidly decreases [19, 20] (Figure 3).

Figure 3: Lift and drag variation with Reynolds number, reproduced from [] In addition to the physical challenges, given by the intrinsic reduction of physical parameters, there is also a problem of system integration. One can easily be misled to believe that larger aircrafts are much more complex than small AVS. The complexity of AVS becomes apparent if it is considered that they, similar to a larger aircraft, should be fully operational with respect to flight altitude, acceleration, stability, speed, and so forth, while the sensors and signal processing units, as illustrated in Figure 4, have to be integrated in a much smaller volume, with limited weight while keeping the power consumption to a minimum, increasing the challenges beyond that of larger aircrafts.

Figure 4: Air Vehicle System (AVS) system integration

LITERATURE REVIEW What is Mini Air Vehicle? Mini air vehicle, or micro air vehicle, or micro aerial vehicle, is a class of miniature UAVs that has a size restriction and may be autonomous. Modern craft can be as small as 5 centimetres. (“Micro air vehicle,” 2018) Development is driven by commercial, research, government, and military purposes. The small aircraft allows remote observation, impossible rescue mission, observation of hazardous places and accessible to a new place that is not accessible by big size human. MAVs have been built as a new hobby purposes to humans as more technology that allow micro system or micro parts to be made precisely from computer design by using 3d printers.

What is Flapping Mechanism? Based on figure 1 above, flapping mechanism is a linkage mechanism, driven by a DC motor, is used to transform the motor shaft rotation into the flapping motion of the wings. It has two stages; a slider crank based mechanism generates low amplitude oscilation motion and a four bar linkage amplifies the motion to desired 120 degrees. The slider is common for two wings.(Karásek, Hua, Nan, Lalami, & Preumont, 2014) However, our research will search for a new gear mechanism that can actually fit our small scales air vehicles. The idea is that to avoid using a free crank and used a hinge instead as it will reduce the size of our mechanism later on.

Kinematics Lift In order for an aircraft to rise into the air, a force must be created that equals or exceeds the force of gravity. This force is called lift. In heavier-than-air craft, lift is created by the flow of air over an airfoil. The shape of an airfoil causes air to flow faster on top than on bottom. The fast flowing air decreases the surrounding air pressure. Because the air pressure is greater below the airfoil than above, a resulting lift force is created. To further understand how an airfoil creates lift, it is necessary to use two important equations of physical science. The pressure variations of flowing air are best represented by Bernoulli's equation. It was derived by Daniel Bernoulli, a Swiss mathematician, to explain the variation in pressure exerted by flowing streams of water. The Bernoulli equation is written as:

The Bernoulli equation states that an increase in velocity leads to a decrease in pressure. Thus the higher the velocity of the flow, the lower the pressure. Air flowing over an airfoil will decrease in pressure. The pressure loss over the top surface is greater than that of the bottom surface. The result is a net pressure force in the upward (positive) direction. This pressure force is lift. There is no predetermined shape for a wing airfoil, it is designed based on the function of the aircraft it will be used for. To aid the design process, engineers use the lift coefficient to measure the amount of lift obtained from a particular airfoil shape. Lift is proportional to dynamic pressure and wing area. The lift equation is written as:

Where S is the quantity in parentheses is a dynamic pressure. For designing Micro Air vehicle, we should get the high coefficient of lift so it can be lift in optimum force

Wing The increased speed over a curved, larger wing area creates a longer path of air. This means the air is moving more quickly over the top surface of the wing, reducing air pressure on the top of the wing and creating lift. Also, the angle of the wing (tilted) deflects air downwards, causing a reaction force in the opposite direction and creating lift. Larger wings produce greater lift than smaller wings. Wing loading tells you how fast a bird or plane must fly to be able to maintain lift:

wing loading = weight/wing area (kilograms per square metre).

So we need to adjust to create wing loading for our MAV have the lightest wing loading A smaller wing loading number means can fly more slowly while still maintaining lift and is more manoeuvrable.

Wing Loading Defined as ratio of the weight of the object and the wing area. The standard values usually chosen are the maximum gross weight and projected area of the wings on a horizontal plane. (“Unsteady aerodynamics and flow control for flapping wing flyers - ScienceDirect,” 2003) Coefficient of lift as stated above. In level flight the lift force equals the body weight, W, and so the wing loading can be expressed as:

Biomimicry

-

Inspiration from Nature

A new trend has started in research community which attract them to look for ideas by looking at nature itself. Our world has a lot of mysteries that still unfolded. Some of them take inspiration from flying insects or birds to achieve unprecedented flight capabilities.(“Micro air vehicle,” 2018) Biomimicry does not only inspired flying, it also has inspired many engineers to make great new things. Take the bullet train in Japan as an example, the Shinkansen, it is inspired by kingfisher bird which penetrate quietly into and out from a tunnel.(“Biomimicry – It’s A Bird, It’s A Train,” 2010) This does show that by taking inspiration from the world, does make our design and invention more creative and great. Therefore, for this research, our group will look at how bat wings can help us achieve greater flight and control but at the same time wonderful to look at.

Scope Limitation 1. Low Quality of the machine and tools: a)3D printer have low resolution when printing. Hence, the 3D model that been printed not so rigid and easily break. 3) tools at the workshop such as hot glue gun is not working so well.

2. Most of the equipment are shipping - Lot of time wasted waiting equipment delivered. Shipping item from oversea required from 2 to 3 week to delivered. Extra charge on the shipping included. - For instance, our dc motor took around 3 to 4 weeks to be delivered. It can affect the working progress to build up the flying mechanism.

CHAPTER 2 FEASIBILITY STUDY/ MARKET SURVEY

Market Analysis

Technavio has published a new report on the global unmanned aerial vehicle market from 2017-2021. (Graphic: Business Wire)

According to the latest market study released by Technavio, the global unmanned aerial vehicle (UAV) market is projected to grow to USD 19.85 billion by 2021, at a CAGR of nearly 13% over the forecast period. Based on the UAV type, the report categorizes the global unmanned aerial vehicle (UAV) market into the following segments:  Medium-altitude long endurance (MALE) UAV 

High-altitude long endurance (HALE) UAV



Tactical UAV



Micro UAV



Mini UAV

As seen in the graphic above, micro UAV only occupy 12.29% of the 2016 market share. This is because the top three revenue-generating UAV segments in the global unmanned aerial vehicle (UAV) market are Medium-altitude long endurance (MALE) UAV, High-altitude long endurance (HALE) UAV and Tactical UAV.

Customer Analysis

Product & Technology Benchmarking We set to benchmark our product with 3 of these mini UAVs’: 1. CIT, AeroVironment and UCLA "MicroBat" ornithopter The MicroBat ornithopter from the California Institute of Technology (Caltech), working with AeroVironment and the University of California, Los Angeles. The ornithopter design concept followed experiments conducted in the mid-1990s by Charles Ellington, a zoologist at the University of Cambridge, and his colleagues, in which mechanical analogues of insect wings were tested in a wind tunnel. The group was only interested in studying the biomechanics of insects and was extremely surprised that somebody seemed interested in them. The CalTech / AeroVironment MicroBat ornithopter was test-flown for short distances under battery power. Researchers performing flight tests with the MicroBat said it tended to attract small birds when it ran low on power and fell to the ground. The birds clustered near the floundering ornithopter in what seemed to be a desire to help. Other research groups also worked on ornithopters. A Georgia Tech Research Institute group built a rubber-band powered entomopter and also did research on a chemically powered Reciprocating Chemical Muscle propulsion system.

2. Lutronix Corporation "Kolibri" micro-helicopter The Kolibri micro-helicopter built by Lutronix Corporation of Del Mar, California. The Kolibri (German for "Hummingbird") was larger than the other DARPA MAV prototypes, with a weight of about 300 grams. The Kolibri was built as a cylinder with rotors at one or both ends, using vanes moved through the rotor airflow by piezoelectric actuators for flight control. It was powered by electric motors or a tiny, highly efficient multi-fuel engine developed by a company named D-STAR.

3. AeroVironment "Black Widow" flying-wing The AeroVironment Black Widow MAV. Developed by a team led by Matt Keenon, the Black Widow was powered by electric motor driving a small propeller in the nose, with a lithium battery permitting about 20 minutes of flight. It carried an off-the-shelf camera chip giving it a colour video resolution of 510 by 492 pixels. While the first Black Widow prototype was a flat disk with a single vertical stabilizer and a propeller in the front, it was followed by an improved Black Widow that looked a little like a thin portable CD player with tapered edges and cut-off corners; a propeller in front; and three fins on the back. It did not have autonomous navigation capabilities, and was controlled essentially like a hobbyist's RC airplane.

CHAPTER 3 CONCEPT GENERATION

Chapter 5 – Concept Generation A- Defining Design Problems Based on the marketing survey, mostly the quality of the Micro Air Vehicle have encountered or used is evaluated as 2 (low).There is some factor that lead to the low quality of MAV’s. 1.

Size Competency

Size competency of the Micro Air Vehicle is one of the main problem in designing the wings,internal parts and its body. Due to this smaller design, the quality of the materials need to be upgraded as its need to fly in surrounding area with varies temperature and speed of air.based on the original (ca. 1995) official Defense Advanced Research Projects Agency (DARPA) Micro Air vehicle def- inition requiring a maximum 15 cm (6 inch) dimension confirmed the name to be a total misnomer” (Michelson, 2006) .It means when the size is smaller it will disrupts the movement of air as it will effect the viscosity of the air towards the wings . Ultimately a micro air vehicle because this represented the juncture at which low Reynolds number effects begin to dominate, and beyond which, integration of energy, propulsion, aerodynamic structures, and intelligence is a necessity.Other than that, if its to small makes teleoperation impractical because a ground station pilot cannot see it beyond 100 meters.Lightweight, low-speed MAVs are often more rigid than larger vehicles due to the non linear scaling of the strength of materials. In the presence of wind gusts, the wing structures are less apt to flex and hence the entire body of the MAV undergoes “heave”. Not only does the control of the MAV become complicated, but the aerodynamics of the air flow over the structure can differ radically from that expected at high Reynolds numbers.

2.

Signal communication & Teleoperation

Other than that, there will be on signal communication of the Micro Air Vehicle will be distrupts as the specification of antenna cannot meet the spefication of the will no connected effectively “A 15 cm MAV can only support a maximum antenna aperture of 15 cm . Depending upon antenna type, this dictates a frequency of operation in the 2GHz range. At this frequency, foliage penetration is difficult and line-of-sight transmissions are necessary. Were a user need to send his MAV only 1 km ahead of his current position to look over the surrounding. An onboard camera allowing the ground pilot to stabilize and navigate the craft was first demonstrated in the Aerovironment Black Widow, but truly micro air vehicles cannot carry onboard transmitters powerful enough to allow for teleoperation. For this reason, some researchers have focused on fully autonomous MAV flight. 3.

Endurability

In order for practical MAVs to be created for either indoor or outdoor applications, they must first be able to fly, be controllable, and have a useful endurance. Key to the ability to fly is an efficient aerodynamic structure with a sufficiently high lift-to-drag ratio that it can support the weight of its structure in flight. Weight and strength of materials are essential elements to the creation of any flying vehicle; however, at a scale of 15 cm or less, the area of the aerodynamic surfaces is limited and the gross weight that can be sustained in flight quickly becomes less dependent on aerodynamic efficiency than on propulsive power. Propulsive power for a useful mission endurance is then a function of the energy density of the fuel carried. Most research to date has focused on aerodynamics and flexible structures, with propulsion (energy) and controllability (actuation) receiving far less attention. In fact, energy storage and actuation techniques are essential to useful MAV designs, as they dictate endurance and mission utility. Closer investigation reveals that one can not simply scale down large designs to the 15 cm scale and below, because the interaction of objects moving through air changes as the size of the objects diminish. Classical aerodynamics used to design airfoils in manned airplanes and helicopters no longer applies as the scale of the airfoil approaches

that of small birds and insects because of the reduction in Reynolds number which describes the behavior of the air as seemingly much more viscous. Reynolds number (Re) is a dimensionless number that relates inertial forces of an object such as an airfoil, to viscous forces in a fluid (air). Thinner airfoils are a typical result of designs optimized for lower Reynolds numbers.

B- Solution Exploration 1.1 Aerodynamics MAVs operate in the low Reynolds number regime below 100000. Characteristics such as the lift-to-drag ratio of a flight vehicle change considerably from those of manned aircraft upon entering the low Reynolds number regime. In particular, flow separation and laminar turbulent transition can result in substantial changes in required airfoil shape in order to maintain or enhance aerodynamic performance. With the breakdown of classical aerodynamic analysis methods, MAVs are often best tested empirically to gain an understanding of their actual behavior. Many factors come into play when trying to optimize MAV designs of all types (fixed, rotary, and flapping) including laminar–turbulent transition, angles of attack, leading edge vortex evolution and progression, airfoil shapes, stall margins, structural flexibility, and time-dependent fluid and structural dynamics of the MAV as a system. In addition, techniques and phenomenology never considered in manned aircraft design can be leveraged for increased flight efficiency including various unsteady lift-enhancement mechanisms such as leading-edge vortex control, wake capture, and clap-and-fling mechanisms. 1.2 Structures and materials MAV structures must be strong, but moreover, lightweight. Because the strength of materials does not scale proportionately as things get smaller, we find that materials that would be otherwise unsuitable for aircraft use at a larger scale can become quite useful at the 15 cm MAV scale and below. For example, aircraft parts that might be created with a fused deposition modeling machine from silicone of the wings would be useful in creating molds for actual flight-worthy aluminum components, but at the scale of a MAV, the fused deposition modeling machine could create flight-ready components out of the same silicone formerly relegated only to model. making. Because these smaller structures are often stiffer at these scales, MAVs can tend to be very rigid unless intentionally made flexible as has been done with interstitial wing materials at the University of Florida (Ifju, et al., 2001). Flexible structures can also be used to eliminate actuators with the concomitant benefit of reduced weight and power consumption. For example, in flapping-wing vehicles, the wings typically rotate about their root to achieve different angles of attach on the up-stroke (thrust) and the down-stroke (lift). Strong actuators that are free to move in the flapping direction as well as create rotation at twice the flapping frequency are required. If instead the wing is able to attain a particular angle of attack based on up-stroke and down-stroke aerodynamic loading, the need for twist actuators is actually obviated through the use of smart materials. This technique is employed in the design of the Entomopter, which does not use any wing actuators other than the prime flapping drive motor (Michelson and Naqvi, 2003). Another aspect of MAV design that is intimately tied to the correct selection of materials is the method of control surface actuation. Most control surface actuators are electromechanical in nature, although other ways exist to effect control (Michelson, 2002). Actuators must be able to move with enough deflection to effect a change in the flow over a control surface while at the same time having sufficient force to do so under all flight conditions. Further, for reasons stability, actuators must have sufficient bandwidth to allow the desired critically damped vehicle response to a commanded input. Various actuation materials have some of these characteristics, but often, not all. In addition, some materials are constrained to work only at high voltages or high currents, which are incompatible with the very limited onboard energy source. Some actuator types that have been considered for use with MAVs include conducting shape memory polyurethane (CSMPU) or shape memory alloys like NITINOL wire (high current demand/bandwidth limited), piezoelectric

materi- als (high voltage/low motion), electroactive polymers (high voltage), rheological fluids (bandwidth/interface issues), or ionic polymeric-metal composites (IPMC) (reasonable bandwidth/low voltage/high motion/short life). There are non-electrical actuation solutions as well, including poly- mer hydrogels, chemically fueled pistons, and pneumatic “air muscles”. As these actuators shrink in size, different tech- nologies present better or worse solutions based on the throw, force, reaction frequency, and actuation energy required. 1.3 Flight control Stability and control of outdoor MAVs is a matter of great concern because there is not enough power, mass, or control surface area on a MAV to fight the extremes of the environment. MAVs can fly very fast in order to pass through local wind gusts without significant offset; however, this makes them less useful in “close-in” reconnaissance. Fixed-wing MAVs are particularly susceptible to roll perturbations, and even if flown by a ground pilot, need roll stability augmentation. Increasing demands are being placed on the hardware and software that comprise MAV guidance and control systems. As MAVs become more autonomous, guidance and control systems must support advanced functions such as automated decision making, obstacle avoidance, target acquisition, target tracking, artificial vision, and interaction with other manned and unmanned systems. While performance requirements are increasing, the acceptable form factors (size and weight) of these systems are decreasing. In some instances, the weight of all onboard electronics (both avion- ics and payload) may be under 10 g. Current miniaturization techniques can accommodate this physical footprint; how- ever, the main challenge continues to be processor speed and storage capacity to allow for an ever increasing degree of onboard intelligence. 1.3.1 Attitude control and navigation At the lowest levels, inner loop flight control is concerned with simply maintaining the vehicle in the correct attitude (roll, pitch, yaw) while maneuvering through environmental perturbations (wind, precipitation). For autonomous flight, it is common to separate the flight control problem into an inner loop that controls attitude and an outer loop that con- trols the translational trajectory of the vehicle. Of particular interest to both the inner and outer loops is the issue of a high stability, or updatable, reference. For the inner loop this is either a gyroscopic reference, integrated accelerometers, or external passive (EO/IR), or active (radar/sonar) cues. For the outer loop, GPS is an ideal reference so long as it is available. Indoors, GPS is unavailable. On the battlefield, it may be denied. For very small MAVs and nano air vehi- cles (NAVs), the ability to carry an efficient GPS antenna is not possible because the vehicle itself falls well below the wavelength aperture of the L1 and L2 GPS frequencies (1575.42 MHz and 1227.60 MHz, respectively, or about λ = 22 cm). Dynamic inversion and neural-network-based adaptations have been used to increase performance of the attitude control systems, and a method called pseudocontrol hedging (PCH) has been used to protect the adaptation process from actuator limits and dynamics. In doing so, adaptation to uncertainty in the attitude, as well as the translational dynamics, is introduced to minimize the effects of flight control model error in all six degrees of freedom, thus leading to more accurate position tracking (Johnson and Kannan, 2005). Optical flow is another control technique that has gained popularity since the advent of MAVs. Researchers have determined that some insects, such as the honey bee, observe the bilateral flow of objects in their field of view in order to assess their speed and trajectory relative to objects (assumed by the bee to be on the ground) (Srinivasan et al., 2004). Advances in software technology have the potential to revolutionize control system design. Component-based architectures encourage flexible “plug-and-play” extensibil- ity and evolution of systems. Distributed object computing allows interoperation. Advances are being made to enable dynamic reconfiguration and evolution of systems while they are still running. Technologies are being developed to allow networked, embedded devices to connect to each other and self-organize (Wills et al., 2001).

These capabilities are known as an open control platform (OCP) architecture that allows complex systems to coordinate distributed interaction among diverse components while supporting dynamic reconfiguration and customization of the components in real time. Its primary goals are to accommo- date rapidly changing application requirements, incorporate of new technology (such as hardware platforms or sensors), and allow undegraded operation in heterogeneous or unpre- dictable and changing environments. Regardless of the flight control methods employed, the end result must be a system capable of responding to the high bandwidth maneuvering requirements posed by the MAV. Especially for indoor flight scenarios, the MAV must be able to sense and react to avoid disaster in an obstaclrich environment. Relative to the speed of flight, the onboard sensors must be able to detect obstacles in enough time for the control system to replan the trajectory and command actuators that are able to follow the commanded path before collision occurs. These dynamics are some of the most challenging for any flight vehicle.

C- Creative thinking method A better engineering approach is to use “biological inspiration”. Using biological inspiration, the designer looks at biological structures in terms of their function, and then figures out how to leverage the physical principles involved, to create a mechanical analog that is not an exact copy, but works with similar principles and is able to be implemented. Implementability is essential to a valid MAV design philosophy. At present, engineers understand how the muscles in the Hawk Moth (Manduca sexta) cause the wing to flap for propulsion, and how they can be phased to create differential flapping for flight control. However, the engineers have yet to create an actuator with the efficiency, weight, elon- gation, and speed of the Hawk Moth muscle and are therefore unable to replicate a freeflying Hawk Moth (see Figure 2) in hardware. On the other hand, engineers have measured and observed the kinematics of the Hawk Moth flapping- wing (Willmott, Ellington and Thomas, 1997; Willmott and Ellington, 1997), and coupling this knowledge with other techniques not found in the animal kingdom, have devised flapping MAVs that can be built using present technology (Michelson, 2002). MAV design philosophies should embrace optimiza- tion of all other subsystems to the greatest degree and then concentrate on maximizing performance for a particular mis- sion by increasing the energy density of the stored onboard fuel source. The design space for increasing energy density is greater (and therefore more flexible) than that of optimizing other parameters such as lift, decreased drag, or strength of materials. Bio inspiration example Principle Component Analysis (PCA). These analyses synthesize the backbone design elements of MAV’s articulated flight mechanism and provide insight to the reference inputs to the motors powering its flapping flight such that MAV mimics the behavior of a biological bat. This procedure requires consideration of past bio-inspired robots that applied fundamentally similar design approaches. The efforts in quantifying complex behavior of biological mechanisms in a lower dimensional subspace has led to the successful design of bio-inspired robots that can mimic their biological counterparts to a great extent [1, 3, 7, 9, 19]. Bio- inspired robots are sometimes designed to take advantage of features in animals that are called muscle synergies, or unified activations within groupings of muscles. This notion of synergies, first proposed by Bernstein [2] is based on the theory that it is very difficult for the central nervous system to independently control all of the joints of an animal simultaneously. Kinematic movements likewise have synergies as a result of these muscular synergies. These synergies often form a set of basis vectors where only the most dominant are needed to approximate the animal’s movement. For example, one Degree of Freedom (DoF) in animals may correspond to the coordinated movement of multiple joints [2]. One DoF is not necessarily expressed only as one joint because often movements of joints are coupled to each other. As a result, many bio-inspired robots have been designed to replicate synergies

found in animals in order to design a robot that accurately replicates the behavior of this animal in spite of retaining fewer DoFs.

Fig. 2: DoFs of a biological bat. In producing this figure, an image from Riskin et al. [15] is used. Developing a bat-size MAV has several challenges and restrictions that have roots in weight, size, and power limitations [10, 11]. These restrictions motivate better understanding and selection of major DoFs in bats. Additionally, bat motion can be described in a low dimensional space using PCA. Riskin et al. [15] found that there are three groupings of joints in a bat wing that move together, accounting for 14 of 20 joints. This study also discovered that approximating the bat’s motion with only one third of the principal components accounted for 95% of the variance of the original behavior. It should also be recognized that a bat wing has a very similar skeletal structure to that of a human hand. Though grasping movement is not usually periodic and bat motion is, the similarity of the bone structures and the success of synergistic design of robotic hands gives an optimistic perspective to this approach for MAV’s flight mechanism. It can be seen that mimicking the kinematics of a biological bat is challenging because of its complex morphology. Implementing a bat’s 40 DoFs as a robot would require a very large number of actuators. Given the strict weight requirements necessary for flapping flight and the current limitations of technology, it is essentially infeasible to do this. Simplifications are therefore required for flight to be possible. A synergistic design approach inspired by functional group joints in biological bats has been used to synthesize a kinematic topology for MAV that has a reduced dimensional complexity but retains similar behavior to the studied organism [14]. The current work expands upon recently published works [12, 13, 14] on nonlinear dynamic modeling, flight control design, and hardware development of MAV by presenting a methodology for optimizing the articulated kinematics of a wingbeat cycle of MAV for this previously designed structure. It relies on the fundamentals of PCA to minimize the differences between the two most dominant principal components of MAV and those of a biological bat. An optimizer minimizes the sum of squared differences between Euclidean positions of corresponding markers on a biological bat and MAV as well as that between eigenvectors from PCA such that MAV tracks the trajectories of the markers on the biological bat and acquires matching synergies. The result of this procedure gives the actuator trajectories over a wingbeat cycle and the parameters describing MAV’s structure. It is worth noting that the resulting kinematics is not guaranteed to yield stable flight dynamics, and thus closed-loop feedback is required [12, 13]. The work in this paper is organized as follows. Section II describes the biological bat motion capture data which will be compared against the kinematic behavior of MAV. The construction and capabilities of MAV are briefly outlined in Section III. In Section IV, a parametric kinematic model of MAV is derived that expresses the markers’ positions

in terms of the optimization variables, i.e., the position of the actuated coor- dinates and the physical parameters of MAV.

D- Concept Sketch

CHAPTER 4 CONCEPT SELECTION

CONCEPT SCREENING Gear transmission selection Transmission style A

Transmission style A involve 2 gears connected to a pinion while the pinion is directly contacted a small motor. A shaft also connected to each motor to a respective wing for the linear movement of flapping. Based on our study from Solidwork simulation, we will have a smooth transmission and the torque deliver between the gears is good. We also have tested its resilience to any external forces onto it, and the study shown that it can adhere with any crashes and not broke if it does happen. A small amount of unit involves in this transmission make it lighter and smaller. Transmission style B

Transmission style B involve 2 unlinked gears connected to a separated to achieve a better stability while a shaft is connected between the gears and the wing to transmit the linear movement for flapping. Based on our study from Solidwork simulation, we will have a not very smooth transmission but the torque deliver between the gears is still good. Because of two different motor used, we will have a not similar flapping thus creating a lot of problem when manoeuvring it. We also have tested its resilience to any external forces onto it, and the study shown that it can adhere with any crashes and not broke if it does happen. A small amount of unit involves in this transmission make it lighter and smaller.

Transmission style C

Transmission style C involve a gears connected to a pinion while the pinion is directly contacted a small motor. A shaft is used as a linkage with the help of slot to allow the movement of up and down of the wing when the gear rotates. Based on our study from Solidwork simulation, we will have a smooth transmission and the torque deliver between the gears is good. However, because of the slots, we will have a lot of losses because of friction and it will cause wear in the future. We also have tested its resilience to any external forces onto it, and the study shown that it can adhere with any crashes and not broke if it does happen. This kind of mechanism require a lot of forces because of it sizes and weight which make it not suitable for a small size drone selection. Wing structure Spring mechanism

Spring mechanism allow a smooth movement of the wing to contract and expand cause by the the leg of our MAV. Spring mechanism also used small number of part thus make it adjustable to be small, lightweight and cheap. The problem might happen is that the loss of stiffness of our spring thus we must ensure that it is replaceable in the future. Suspension mechanism Suspension mechanism allow the movement of the wing to contract and expand cause by the

movement of leg of our MAV. To allow the movement to be smooth and efficient, we need to use a lot of part for the perfect placement of our suspension and the tip of the wing. More parts involve thus create a lot of weight thus make it heavier, expensive and not suitable for a small MAV. Suspension used springs, thus potential for it to lose its stiffness can happen. Thus we must ensure that the spring is replaceable in the future.

Wing material

Plastics [Elaborate the stuff above based on concept scoring below. Example transmission style A] Plastics bag is lightweight, shape able and easy to find. It also will allow a smooth movement of the wing to contract or expand according to the leg of the MAV. However, based on our customer survey, they need a MAV that can sustained from any accident if happen while controlling the MAV. Thus making plastics bag very low score in Resilience part of scoring.

Isoprene [Elaborate the stuff above based on concept scoring below. Example transmission style A] Isoprene sheet is lightweight, shape able and easy to find. It also will allow a smooth movement of the wing to contract or expand according to the leg of the MAV. Plus, based on our customer survey, Isoprene elastic characteristic thus making the wing more durable and not easily tear if any accident happens. However, the cost of isoprene membrane is slightly expansive compare to plastics. Bat leg mechanism Rack and pinion

Rack and pinion allow the leg of our MAV to move left and right based on our command from controller. The advantage of using this rack and pinion is that allow greater assist or greater power to the leg. However, for MAV use, the power is not so much needed. Only a slight movement will allow the wing to contract or expand based on the desired of the user because the weight of the wing is not heavy. Rack and pinion use a lot of parts and thus makes it heavier and expensive for manufacturing it.

Servo Motor Servo motor helps the leg of our MAV to move left and right. Allow the wing to contract and expand based on controller by user. Servo motor is just one-part compare to rack and pinion. It will connect directly to AM receiver thus makes it lighter and less expensive. The problem with servo motor is that the size itself. Our team must find a way to fit it into our MAV.

CONCEPT SCORING 1 – Very Poor/Unlikely

Transmission selection

Wing Structure

Wing Material Bat Leg Mechanism

3 – Average

5 – Very Good/Very Much Agree

Low Cost

Smooth transmission

Transmission style A Transmission style B Transmission style C Spring Mechanism Suspension Mechanism Plastics

3

5

4

3

2

Isoprene Rack and pinion Servo Motor

High torque delivery 4

Resilience

Light/Not heavy

Minimum parts involve 4

Small Size Achievability

Total Score

5

34

4

5

3

4

3

3

3

27

1

1

2

1

1

1

11

5

5

4

5

5

5

5

39

2

4

4

3

2

1

2

23

5

3

2

2

4

5

5

28

2 2

5 2

5 4

5 3

3 3

5 2

5 2

35 20

4

5

5

5

3

5

3

31

CONCEPT ANALYSIS Finalize Concept Design Selection As a result of detailed analysis, concept __ was chosen to be the most compatible with customer’s needs. Therefore, our team will proceed with concept ___ for our 2nd semester.

Concept ___: Transmission style A + Spring mechanism + Isoprene + Servo motor