IARC Team Quadrotor, Notas de estudo de Engenharia Mecânica

Baixe IARC Team Quadrotor e outras Notas de estudo em PDF para Engenharia Mecânica, somente na Docsity!

i

Nate Carlos Ben Cole John Cook Jonathan Forest Sansen Johnson Ed Massie Chris Rogers

IARC Team Quadrotor

Table of Contents

1.0 Introduction For 2009, the International Aerial Robotics Competition will be holding its 5 th^ mission since its inception nineteen years ago. Virginia Tech’s Department of Aerospace Engineering created a team of students to compete in this year’s competition. This group paired together with Mechanical Engineering students to comprise the complete Virginia Tech team that hoped to compete in the IARC competition this year. However, due to the difficulty of the design problem presented by this competition, this group will not compete in this year’s IARC competition. The Mechanical Engineering students were responsible for navigation and target acquisition, which includes sensor integration and processing, while the Aerospace Engineering students were responsible for the design selection, creation, and testing of the aerial vehicle that would have been utilized in the competition. This report will cover all portions of the project up to its current state. The team identified four distinct aspects that were vital to successful completion of this year’s mission. The four sub-groups of this year’s IARC team were finance, propulsion, weights/structures, and flight controls. In the following sections, each sub-group will define their unique responsibilities, as well as the progress made in this academic year. This section will cover all aspects of the competition. The competition requires that each team construct a fully autonomous aerial robot that is able to fly and navigate in a confined environment, specifically indoors. Teams are given four flight attempts. Initially, the competition vehicle (CV) will be launched from an area 3 meters away from the building. The CV will then be required to enter the target building through a one (1) square-meter opening. Once inside, the CV will have ten minutes to search, locate, and photograph a target. This information will then be transmitted back to a ground station. Obstacles such as columns, furniture, and interior walls will not be disclosed before the vehicle enters the building; thus, the CV must navigate autonomously. Upon successful completion of the mission, CV is not required to exit the building. The target of interest will be a gauge displaying specific values. The mission is deemed a successful one if the judges are able to read a specific value on the gauge. The gauge will be surrounded by various blinking lights as well as an audible warning tone. The gauge of interest will have one non-blinking blue LED directly below it. It will be necessary for the CV to indicate that it has locked onto the target gauge by means of a JAUS-compatible message as well as a minimum of five seconds of continuous video relayed by a radio frequency. This transmission will need to have enough power to be received at one-hundred meters with a maximum loss of 6

dB. After successfully capturing the picture, the image must be relayed with enough power to be received at one-hundred meters while having a loss of only 6 dB. The receiving antenna can have an aperture no greater than one meter. In addition, the CV will be required to supply information and mission status. These include things such as navigation information, obstacle locations, as well as target location and lock status. JAUS protocol must be utilized during communication with the ground station. The rules below are the main guidelines for the competition as they were written in the published rules [1]:

1. Vehicles must be unmanned and autonomous. They may be intelligent or preprogrammed, but they must not be flown by a remote human operator.
2. Computational power need not be carried be the air vehicle. Computers may be set up in the designated ground station area, but there can be no human intervention with any ground-based systems necessary for autonomous operation.
3. Data links will be means of radio frequencies in any legal band for the location of the arena. The vehicle must be tolerant to any and all interference that it may encounter during the mission.
4. The air vehicle(s) must be free-flying, autonomous, and have no entangling encumbrances such as tethers. During flight, the maximum dimension of the air vehicle shall not exceed one (1) meter. The maximum takeoff weight of the vehicle shall not exceed 1.50 kg. The vehicle must be powered by means of an electric motor using a battery, capacitor, or fuel cell as a source of energy.
5. Each vehicle must be equipped with a method of manual override of the propulsion system to function as flight termination. This override must be capable of operation even if all flight control system function is removed. The operation of this override will be controlled by the Judges.
6. A maximum of two (2) non-line-of-sight (NLOS) navigation aids may be used external to the designated flight area. The navigation aids must be portable, and must be removed once the team leaves the competition area. GPS, GLONASS, Galileo, or other satellite navigation systems are not allowed as navigation aids.

2.0 Preliminary Design Selection The mission objectives for this year’s competition posed some interesting difficulties. The first challenge is the enclosed environment in which the vehicle will be required to operate. Given that the vehicle will spend most of its time indoors, a conventional fixed wing design did not seem appropriate given the higher forward speed required to maintain enough lift for flight. Additionally, the vehicle will be required to video and photograph a stationary position on a wall, and this favors a platform that has the ability to hover. With this in mind, the team produced a list of possible designs. The initial design concept list is below:

• Single Ducted Fan
• Single Ducted Fan with two side fans
• Dual Ducted Fan
• Lighter than Air
• Conventional Helicopter
• Coaxial Counter-Rotating Main Blade Helicopter
• Quad Rotor The following pages will show the initial design concept sketches for each possible design case. Then, the team’s design criteria will be introduced and explained. Finally, each proposed design will be analyzed, leading to the design that was selected for this year’s competition.

Figure 2: Single Ducted Fan

Figure 3: Single Ducted Fan with Two Fan Sides

Figure 6: Conventional Helicopter

Figure 7: Coaxial Counter-Rotating Main Blade Helicopter

Figure 8: Quad Rotor The team created an initial design criterion, which was established to help determine which design would accomplish the mission goals most effectively. There were six characteristics to this criterion: inherent stability assessment, controllability, useful payload, durability, vehicle gross weight, and maximum linear dimension. For ease and simplicity, a basic grading scale was created to assist the group in identifying which concept would be the best choice for the competition. There were three possible grades and respective point values that could be assigned. The first was a grade of ‘undesirable’ which would earn the concept a score of ‘-1’. The next was a grade of ‘neutral’ which would result in a score of ‘0’. Finally, was the grade of ‘desirable’ with a score of ‘1’. The inherent stability assessment was defined as the difference between dynamic stability and dynamic instability. For this the team analyzed each case for the location of the thrust point in relation to the center of gravity. If the center of gravity was above the thrust point for the vehicle, it was defined to have an unfavorable inherent stability characteristic. An example of this was the Single Ducted Fan. For this initial design, the thrust point was considered to be the directional vanes on the very bottom of the vehicle. Given its weight distribution, the center of gravity it logically located above the thrust point giving it undesirable inherent stability with regard to accomplishing the defined mission.

Table 1: Down Selection Matrix

DESIGN CONCEPT STABILITY CONTROLABILITY PAYLOAD DURABILITY WEIGHT^ GROSS DIMENSION^ MAX TOTAL SINGLE DUCTED FAN -1 -1 -1 1 -1 1 - SIDE DUCTED FAN WITH 2 SIDE FANS -1 0 -1 1 -1 1 - DUAL DUCTED FAN 0 0 -1 1 -1 1 0 LIGHTER THAN AIR 1 1 -1 0 1 -1 1 CONVENTIONA L HELICOPTER 0 -1 0 -1 0 0 - COUNTER- ROTATING MAIN BLADE HELO 0 0 0 -1 0 0 - QUAD ROTOR 1 0 1 1 1 1 5

3.0 Finances For the second phase of the project, spending decreased considerably as compared to how it was when the project was first taken. This can simply be attributed to the fact that so much was learned in the development of the first prototype in phase one that development of the second prototype was more focused and purposeful. Less money was spent towards the goal of simply figuring out what works. Spending was also reigned in through the reuse of parts from Proto1 in the development of the new carbon fiber model. Another positive was the increase in financial resources, made possible by a donation of $4000 by the Virginia Center for Autonomous Systems. A record of the team’s financial activity was kept in excel sheets as before. Examples of record keeping can be found in Appendices A, B, and C. Appendix A represents a raw accountability for parts purchased, showing reference information such as part number, date of purchase, retail host, etc. Appendix B represents spending distribution across each discipline within the team, while Appendix C represents a cost analysis of the prototype as constructed.

4.0 Weights and Structures The weights and structures subgroup was assigned with designing the overall structure of the quadrotor and keeping track of the actual and estimated weights. The overall placement and layout of components, as well as stress and deflection analysis were under the responsibility of this subgroup. In order to meet these responsibilities, the weights and structures subgroup researched and analyzed other similar quadrotor designs, analyzed various building materials and worked on building an early prototype test platform. Additionally, tables of actual and estimated weights were kept and updated as the various designs progressed. 4.1 Quadrotor Design To meet the competition mission objectives, several design goals were set. It was decided that the quadrotor design was to have no single dimension greater than 0.8 meters in length so as to allow the vehicle to fit through the required 1 meter by 1 meter window while allowing some room for error. To meet the mission objective of having the vehicle weigh less than 1.5 kg, it was also decided to construct the quadrotor out of as many light weight materials as possible, as well as utilizing a minimal amount of structure to reduce weight and complexity. Passive stability was also of concern. To help maximize stability, it was decided to design the quadrotor in such a way as to keep the center of gravity as low as possible. Finally, the weights and structures subgroup set the goal of keeping the quadrotor design relatively simple to construct and repair. This final constraint was implemented so that as much time as possible could be devoted to testing and improving the vehicle, rather than assembling and fixing it. In the beginning design stages, several comparable quadrotor designs were researched and analyzed. Two of the most influential designs were the BYU Quadrotor[2] and the Draganflyer X Pro[3], as seen in Figure 9.

Figure 9: Design influences. At left BYU Quadrotor design[2], right photo Draganflyer X Pro [3].

though it would incorporate more advanced materials and custom parts as it was intended to be the design used in competition. It was originally intended to have Proto 2 built and ready to test early enough that a second carbon fiber design could be built or updated if needed. Delays in its construction led to only one carbon fiber model being built. 4.1.1 Proto The design of Proto1 can be seen in Figure 10. While not designed to be used in competition, Proto1 was designed to be cheap, quick and easy to build, using mostly over-the- counter structural materials. As such, the main spars were to be made of basswood, the base plate made of maple and the ME team’s sensor payload was to be represented by a wooden box filled with sand. Proto1 was mainly to be used as a proof of concept of several ideas. It was believed that building a simpler design of the quad rotor early in the design process would allow for a better understanding of how many of the parts would fit together, as well as giving ideas on how to improve future designs. Additionally, having a cheap, quick-to-build prototype would allow for other subgroups to test their systems on a quadrotor platform similar in design and weight to what was going to be used in competition. In Figure 10, one can see the overall layout and primary dimensions of Proto1’s design. Most of the components were color coded by function or type for easy recognition. The main structure of the quad rotor was colored in shades of blue and black, and included the base plate, main motor spars, and payload connectors. The propulsion system included the motors and propellers and was colored green, while the batteries, electronic speed controllers and other associated electronic equipment was color coded in shades of orange and red. The ME payload was represented by the large purple box.

Figure 10: Proto1 design as seen in Autodesk Inventor 2009. Dimensions in millimeters.

Figure 11: Proto2 as seen in Autodesk Inventor 2009. All dimensions in millimeters.

The Proto2 used the same color coding system as found in the design of Proto1 in Figure

10. From this figure, one may notice that the Proto2 design called for a few structures not seen on Proto1. The shrouds seen at the ends of the main spars had the express purpose of protecting the propeller blades from damage should the quadrotor hit a wall during autonomous flight. Additionally, one may have noticed the outer cross beam spars connected near the end of each of the main motor spars. These outer spars were implemented to reduce the twisting effects felt by the main spars while the propellers were changing speeds. It was determined that these additional spars would be necessary due to the main carbon fiber spar’s smaller cross sectional area and thus reduced torsion al stiffness. Additionally, it was felt that it would be beneficial from a flight control’s stand point to keep the motors on a spar mount as structurally stiff as possible. As carbon fiber was found to be much stronger than basswood, the main spars chosen for Proto2 had a much smaller cross section of 4.78 mm x 4.78 mm with an inside circular diameter of 3.05 mm (0.188 in x 0.188 in x 0.120 in). While this smaller cross section was able to save weight, it prevented the motors from being able to be attached directly onto the spars as was seen in the design of Proto1. As such, small mounts and brackets were designed and had to be custom manufactured for Proto2. 4.2 Materials An important factor in the design of the quadrotor, both Proto1 and Proto2, were the materials used. Three primary factors were considered when choosing the materials: strength, deflection and weight. Using these criteria, several different materials were studied and compared, including aluminum 6061, G-AIMg5 fiberglass, balsa wood class IV, birch wood class IV, PVC (hard), Styrene (plastic), carbon fiber, and basswood. Limitations of the main motor spars were the primary focus of the initial calculations, as it was determined that these parts would be experiencing the highest weights and moment out of all the parts in the quad rotor. To analyze the various materials, the bending stress and factors of safety were determined using (1) and (2) [4].

I x

σ = My

Bending stress