UAV Club News
27 August 2008
Team attends 2008 competition
In July, the team attends the IARC competition and places 5th out of 10 teams attending. The team brings back $5K in prize money plus an award for most innovative design. Gallery:
1 June 2007
Report posted for 2008 competition
Paper submitted for 2008 competition available here.
1 June 2008
Subscale model of Interior Search Vehicle (ISV) mated to Drop Vehicle. The ISV consists of the aft extension of the Drop Vehicle fuselage. The ISV is ejected after the Drop Vehicle enters the building.
5 May 2008
Upgraded hovering vehicle
Upgraded version of hovering vehicle was tested on 2008/3/28. This test was under manual control. Passive stability is used for roll and pitch axes. No gyros or accelerometers are used. The balloon stabilizer in the previous version was replaced by an inverted umbrella stabilizer.
3 January 2008
Experimental hovering vehicle was tested on 2007/12/31. The vehicle is intended for Levels 3 and 4. This test was under manual control.
10 December 2007
Happy Birthday Ada Lovelace
On December 10, 1815, Augusta Ada Byron was born in London, England.
28 November 2007
Successful 2D and 3D antenna tracking of airplane;
On 18 November, we did a ground test of the dish antenna. While the airplane was being moved around on the ground, the antenna was able to track the vehicle in real time. We also did two short, unpowered glide tests of the airplane to verify that the ground station was able to receive telemetry from the airplane in flight. These were essentially 2D tests of antenna tracking.
On 27 November, we extended the tracking to 3D. While the airplane was in powered flight, the antenna was able to track the vehicle at altitude.
These antenna tests were for tracking only. The antenna itself was not yet functional, since a feed antenna had not been installed yet. A separate 900 MHz radio modem was used to transmit GPS data from the airplane. A ground-based computer received the GPS data and extracted position information, which was then fed to the antenna controller in order to steer the antenna.
We are also beginning to transition our software to Ada. The 18 November tests inaugurated our new ground station software, written in Ada 95.
11 September 2007
Panorama of IARC 2007
A super-wide-angle panorama of the flight line was added to our IARC 2007 gallery.
3 August 2007
Gallery -- IARC 2007
Gallery from IARC 2007, held 23 to 26 July 2007 at Fort Benning, Georgia.
6 July 2007
High gain antenna nearing completion
The steerable high gain antenna is under construction and nearing completion. The antenna controller, based on a BX-24 microcontroller, is able to control both the azimuth and elevation step motors.
6 July 2007
Construction of SVC 55 Pusher begins
Construction of 55 % scale SVC Pusher begins 3 July. Foam blanks have been cut for the airframe base shown below, as of 4 July. The motor is being moved from the nose to the tail, which affects the overall center of mass of the vehicle. The nose is being extended to compensate.
All digits stayed intact this time around (see also).
9 June 2007
Drive for high gain antenna
The design for the drive system for the steerable antenna is nearing completion. Two step motors will be used to control azimuth and elevation:
Click here to interactively rotate the assembly. Adobe Reader V7.0.9 or higher is required to view this file.
1 June 2007
Report posted for 2007 competition
Paper submitted for 2007 competition available here.
12 May 2007
High gain antenna
A cooking scoop is being investigated as a parabolic reflector for a high gain antenna. The antenna will download image data from the primary air vehicle. Range tests of the antenna were made on 11 May by using a small wireless video camera mounted at the end of a 306 cm plastic pole. Tim got the idea of using a cooking scoop from here.
Wesley (below, left) is holding the camera pole that was used for range testing. The cooking scoop is on the right. For the tests, a small handheld video receiver (not shown) was held at the approximate location of the focus. The receiver was moved around until we got the best signal.
Also, a rapid prototype of the radome was built on 8 May. The radome is intended to protect the high-gain antenna from wind loads.
7 April 2007
First flight of experimental camera
Our experimental high-resolution camera was flown for the first time.
20 March 2007
AIAA approves additional funding
On 12 March the AIAA Tucson Section approves $550 in additional funding for the team, which brings AIAA's total to $1050 since July 2006. The funds are greatly appreciated.
16 March 2007
Camera modified for external shutter and power control
Dave modifies the Sony camera so the shutter and power can be controlled by an external computer. Due to the very small sizes of components, a microscope is required for soldering internal electrical connections in the camera.
15 Mar 2007
A reminder that safety has priority. News of construction of SVC 100 in 2004 was updated to include information about a bandsaw accident that occurred at the time.
2 Mar 2007
More camera guts
The Sony camera was dissected further, and Dave used a scope to analyze various signal and control lines. Shown below is the Memory Stick interface. Click for a larger image that also shows the two gyro chips used in the image stabilization system.
28 Feb 2007
New team member
Welcome to new member Chris King, who is a PCC student. Chris has several years of operational UAV experience, both in the Army and as a civilian. Chris was involved with Hunter and Shadow UAVs, and participated in the IOT&E (Initial Operation Test and Evaluation) for Shadow.
17 Feb 2007
Experimental camera guts
A Sony DSC-T10 camera was partially disassembled in order to determine how easy it would be to modify for computer control. The plan is to experiment with the camera to get empirical data we can use to help us design a reconnaissance camera system.
New team members
Welcome to our newest team members:
The students are from the ECE Department (Electrical and Computer Engineering) at the University of Arizona, and are taking the ECE 498A capstone senior design course. The group will be working on our camera system as their class project.
30 Jan 2007
System architecture V1.6
The proposed system architecture is updated. The main change is to use a single air vehicle for ingress, which is similar to our 2005 design. The new design uses a single primary UAV that also carries two subvehicles. All three vehicles are combined together as a single unit for the 3 km ingress.
The previous proposal (V1.5) used dual UAVs to fly in parallel for ingress, which split functions between the two vehicles. One UAV was specialized for reconnaissance and the other was specialized for carrying subvehicles.
The change is made as a result of practical experiences at previous IARC competitions, plus comments made by one of the judges at the 2006 IARC, who was somewhat leery of the multiple-UAV approach for ingress.
22 December 2006
Real time symbol tracking in lab. See
video here (410 KB).
2 December 2006
Test of Ground Rover Testbed
On 29 November we ran ground tests of our new rover testbed. The purpose of the vehicle is to test software and electronics that will eventually be incorporated into the ground rover. Tom Stutsman (below) built the rapid prototype vehicle and got it working in a very short period of time: See rover test log for details.
7 November 2006
Drop Vehicle Testing
The Drop Vehicle was modified by repositioning the control surfaces from the canards to the rear wings for modularity reasons. A Cirrus Micro Joule control system was added, with servos on each of the three rear wings.
The first flight with the new configuration was late on 5 November. Additional flights were made on 6 November. All flights were in a courtyard at Pima West Campus. From the 6 Nov tests, a sequence of frames was extracted from video and is shown below. The flights were under manual control. Control response was good.
27 September 2006
New GPS data recorder flown
A new GPS data recorder was flown. GPS data was downloaded after flight. See flight test log 25 for details.
10 September 2006
Better waypoint tracking, plus new elevon mixer
This was one of our best GPS waypoint tracking tests yet. On 8 September, we tested a triangular waypoint pattern that was smaller than the previous square pattern. Also we tested a new electronic elevon mixer that seems to have solved the elevon glitches we had been seeing. See flight test log 24 for details.
18 August 2006
Waypoint tracking accuracy boosted
Test 52 was repeated except that aileron gain was reduced. Tracking accuracy was noticeably improved. See flight test log 23 for details.
17 August 2006
Successful tracking of square waypoint pattern
A 200 m square waypoint pattern was successfully tracked by SVC 55 on 16 August. See flight test log 22 for details.
2 August 2006
2006 IARC photos posted
Photos are posted of the 2006 IARC competition held on 27 July.
19 July 2006
AIAA approves travel funding
At their council meeting on 11 July, the AIAA Tucson Section approves $500 for travel funding. The funds are greatly appreciated.
19 July 2006
Bent motor bracket
During recent repairs of our primary UAV, we discovered that the motor bracket had been bent. The bracket is used to attach the motor to the motor mount, and damage to the bracket could cause a misalignment in the thrust line. It's possible this could explain the pitch trim problems we had been seeing in the past few flight tests.
The bracket had been in the last crash of 6 July, so it is possible the crash itself caused the damage. So we can't say for sure that the bracket caused the pitch trim problems, but it look suspicious.
6 July 2006
Waypoint tracking test ends in crash
We attempted to track two GPS wayoints 350 m apart. Tracking is not successful, and the flight ends in a crash with relatively minor damage. Check out flight test log 21 for more details.
4 July 2006
Successful tracking of GPS waypoint
On July 4th, in our second set of tests of the new autopilot, the airplane was able to successfully track a GPS waypoint while holding constant altitude. See flight test log 19 for more details.
Also welcome to new team member Tikwiza Botha, who is majoring in Computer Engineering at the University of Arizona.
30 June 2006
New autopilot operational
Two test flights were flown with a new autopilot and GPS receiver. The autopilot now includes adjustable gains on both the aileron and throttle controllers. Reducing the aileron gain seems to have fixed the roll oscillation problem we were seeing previously. Also, a broken wire was found and corrected in the power supply to the control system. See flight test log 18 for more details.
27 June 2006
Added more photos of Globe High School flight demo
We added more photos of the GHS flight demo, plus a link to the GHS site. See updated flight test log 16.
24 June 2006
New autopilot arrives
Our new autopilot and lighter-weight GPS receiver arrived today. The plan is to use the devices as spares and for breadboard experiments in the lab. The autopilot is a Picopilot from Unav.
17 June 2006
As part of the portal detection process, it may be necessary to determine if a windowpane or similar material covers a given portal. We've been planning on using light polarization to assist in this process. One limitation to this approach is that internal surfaces can sometimes generate polarized light, thus mimicing a windowpane.
We are experimenting with a method of discriminating between polarized light reflected from a windowpane and polarized light reflected from selected internal surfaces, such as walls and floors, in the absence of a windowpane. Shown below are portions of two images recorded with different polarization filters. You can see that the variation in windowpane brightness is opposite in polarity compared to the wall and floor.
7 June 2006
New team member
Welcome to new team member Andrew Wagoner, who is starting at Pima College this summer.
1 June 2006
Report posted for 2006 competition
Paper submitted for 2006 competition available here.
28 May 2006
Dry run at Parker Canyon Lake
Dry run at Parker Canyon Lake. Our camera test airplane was taken to a joint campout with AIAA Tucson Section (American Institute of Aeronautics and Astronautics), Tucson Amateur Astronomy Association, and the Raytheon Astronomy and Telescope Making Club.
25 May 2006
System architecture V1.5
The system architecture was changed as follows:
10 May 2006
Demonstration for Globe High School students
On 10 May we did a short flight demonstration for a small number of Globe High School students. SVC 41 was flown under manual control.
20 April 2006
Orbit animation posted
We posted animated GIF files that display short video sequences from flight tests of 7 April. The tests were of our new dual camera system. An orbit was attempted around a test building on which the IARC symbol was attached.
For the dual camera sequence, note that upper and lower frames were not simultaneous. Frame capture times were probably evenly spaced between the two cameras. The average frame rate was approximately 8.62 frame/s (4.31 frame/s per camera).
There are obvious discontinuities between the stacked images from the upper and lower cameras. In principle it would be possible to synchronize the cameras and stitch the two images together, but it's not clear yet whether it would be worth the trouble. A lot depends on the exact procedure we use for the symbol search and portal search.
10 April 2006
Improved steering of 16 % subscale SVC.
A profile of the Drop Vehicle was added to the 16 % subscale SVC in order to increase rolling moment as a function of sideslip angle. Steering was somewhat improved over previous tests in a throttle-only control configuration.
7 April 2006
First flight test of dual camera system
The dual camera system was flight tested for the first time on 7 April. The two cameras are side looking, with adjacent over/under fields of view with a slight overlap. We also got more experience flying the airplane with the mid and aft fuselage removed.
24 March 2006
Camera test with IARC symbol on building
This was a camera test with the IARC symbol attached to building. Side-looking camera pitch angle was about 30 deg down, compared to previous tests at about 45 deg down.
30 March 2006
First powered flight of 16 % SVC
The 16 % subscale SVC was flown for the first time under power. The airplane was also modified by replacing the single tractor propeller with twin pusher propellers.
1 March 2006
VRML test models refined
VRML models of two test buildings have been refined by correcting relative locations of building components, adding a bitmap of one of the roofs, and adding a bitmap of the ground based on an image derived from Google Earth.
10 February 2006
Tour of Physics Lab by 8th graders
Groups of 8th graders from local schools toured the Physics Lab at Pima West Campus on 10 February. As part of the tour, the students were shown air vehicles of the UAV Club. More photos.
20 January 2006
Fourth camera test
Video was successfully recorded from our fourth camera flight test. Stereo pairs were extracted manually, and we're also experimenting with comparisons of video frames with manually-created VRML models of a pair of test buildings.
29 December 2005
System architecture V1.4
The system design was changed to define the symbol search area to a circle of 200 m radius rather than 200 m diameter.
28 December 2005
New team member
Welcome to new team member Natalie Wagoner. Nat helped out with our third camera flight test on 23 December.
28 December 2005
Third camera test successful
On 23 December, our third camera flight test resulted in successful recording of downloaded video. Stereo pairs were extracted from the video. The flight also verified recent repairs and modifications of the airframe.
21 December 2005
Second camera test ends in crash
On 2 December, the second test of the camera ended in a crash, probably due to a loose elevon control horn. A few hours of labor repaired the damage.
8 December 2005
Initial camera tests
On 18 November, a low resolution, wireless video camera was flight tested on the 41 % subscale SVC. The camera was side-looking and pointed approximately 45 deg below horizontal. The flight lasted 8.5 minutes. The camera was a Swann MicroCam III CMOS camera transmitting on 2.45 GHz.
Due to a late start, the flight didn't start until about 17:49 local time, which was well after sunset, and the darkness made it hard to see much detail in the raw video. A limited amount of image enhancement was done on the frames shown above in order to bring out more detail. Enhancement consisted of increasing brightness, increasing contrast and applying a low-pass filter.
Although the camera has a reasonably high frame rate, it has a low resolution. So the eventual plan is to use this or similar cameras for guidance of the drop vehicle rather than for symbol and portal searches. Higher resolution cameras are planned for the latter searches.
30 October 2005
First flight of 16 percent subscale SVC
First flight of 16 % subscale SVC. The primary purpose of the vehicle is for manually-controlled flight demonstrations in small areas, such as parking lots or gyms.
26 October 2005
Welcome to new team members
Welcome to new team members who have joined since the beginning of this semester:
24 October 2005
Conceptual design for ground rover
New Rover concept -- the existing Drop Vehicle (DRV) design is modified by splitting it into separate modules, one of which is specialized as a ground rover. After landing, the aft wing section would be jettisoned, and the canards and forward wing sections would double as wheel spokes on the Rover.
On each wheel, spring-loaded flat panels are folded flush during flight. After landing the panels pop out and form nine-sided wheels. The two wheels rotate independently for steering.
To increase modularity and simplify the wheel design, all flight control surfaces are removed from the canards, and the rear control surfaces are enlarged to compensate. Flight controls would be limited to the rear wing section, which separates from the Rover after landing.
We also considered wheels with inflatable tires (above), but the added weight and complexity of the inflation mechanism was of concern.
In the original 2004/05 conceptual design of the DRV, the ground rover was contained in an internal cylinder that would be ejected from the front of the DRV on landing.
20 October 2005
System architecture V1.3
The system design is modified to classify the second airplane (Carrier) as a subvehicle instead of primary vehicle. Also, Carrier is launched autonomously instead of manually.
18 October 2005
System architecture V1.2
The overall 2006 system design is briefly summarized here. The main change from 2005 is the addition of a second primary UAV. One UAV handles reconnaissance and the second UAV carries the drop vehicle.
8 October 2005
Demonstration of fiber spectrometer
Team member Chien-Wei Han demonstrates a fiber spectrometer in the optics lab at Pima College. The device measures light intensity as a function of wavelength, and can be used to analyze light reflected from a surface as a means of identifying the composition of the surface. The device is light enough to carry on a small air vehicle, and might be helpful in identifying closed vs. open portals at the IARC competition.
27 September 2005
Team competes at 2005 IARC
The PCC team competes at the 2005 IARC competition in Ft. Benning, Georgia on 21 July. For a more details, check out page 3 of the August newsletter of the AIAA Tucson Section.
14 June 2005
Aft ejection of subscale Drop Vehicle
In an effort to reduce pitch and yaw transients of the drop vehicle after release, the release mechanism was redesigned to eject the drop vehicle backwards.
Original version -- the original version of the release mechanism can be seen in some simple wind tunnel testing. Here a rubber band is used to strap the drop vehicle to the top of the carrier UAV. A servo releases the strap. Jason Lewis and Gerardo Cornejo handled the wind tunnel testing.
Wind tunnel testing of original
mechanism (video here)
New version -- in the next iteration, Jason redesigned the system so that a a rail was added to the underside of the fuselage of the drop vehicle. The rail fitted in a slot on top of the carrier UAV, and a spring ejected the drop vehicle backwards relative to the UAV. Video can be downloaded. Individual frames of the separation were also extracted.
Subscale separation test of new release mechanism
The subscale separation tests of 10 June and 14 June were done with a new motor installed in the UAV. The motor is a Hacker A20-34S brushless motor, which has significantly higher power than the previous GWS carbon brush motor. The higher power seems to have solved most of the performance and control problems we were seeing previously. Now the UAV flies beautifully both with and without the drop vehicle attached.
Landing approach after separation test
10 June 2005
Drop Vehicle separation tests
Today we did in-flight separation tests of the subscale drop vehicle. Video of the best separation can be downloaded here. There continues to be a pitchup tendency just after separation, although the pitchup is not as pronounced as it was in previous tests on a different airplane. One surprise was a right yaw tendency by the drop vehicle just after launch, possibly because the carrier vehicle was flying at a nonzero sideslip angle.
Thanks to Jason Lewis for building the release mechanism, as well as Gerardo Cornejo for recording the flight video.
Tail catch -- on one test, the drop vehicle yawed to the right and caught on the right-hand vertical tail (see video. and extracted frames of the tail catch. The drop vehicle eventually freed itself after a couple of seconds.
17 May 2005
First powered flight
The first powered flight of our primary air vehicle took place on 17 May. Control was manual. Pitch control was good, roll control needed more sensitivity and power was more than adequate. The vehicle never needed more than 50 percent throttle.
Approximately 24 seconds into the flight, the aft fuselage separated, taking the R/C receiver with it. Loss of control resulted in a crash, with the vehicle impacting the ground at a near-vertical attitude.
Damage turned out to be relatively light and was mostly confined to the forward and mid fuselage and radio. The motor controller was damaged and the gearbox output shaft was bent, but the motor itself appeared to be OK. One cell in the motor battery was slightly dented. Very little damage occured to the wing, vertical tails or aft fuselage.
16 April 2005
McKenna math model
A VRML model of McKenna was downloaded. The plan is to use the model for machine vision simulations.
In the illustrations shown below, renderings of the model (right) can be compared with camera images recorded by the University of Arizona team at the 2004 IARC competition.
Comparison of camera image vs. math model, as seen from similar viewpoints
Similar comparison, with target building in center
The VRML model was downloaded from the SRI International web site. The model doesn't include a few recent changes to the buildings, but it should be good enough to define a search strategy.
The intent is to use the model solely for simulations and similar purposes, and not use it during the actual competition. It is assumed that accurate GPS building locations are not known beforehand, during the competition itself, if our interpretation of the IARC FAQ is correct. From question 1, the target building is assumed to be located within 100 m CEP of a specified GPS waypoint. The waypoint is assumed to be unknown until the beginning of a run.
13 April 2005
Flight testing started for training vehicles
Flight testing started for SuperFlea training vehicle. The SuperFlea has a configuration that is very similar to our main air vehicle platform -- delta wing, twin vertical tail, propeller driven, elevon control, with skid landing.
SuperFlea in flight, with Tucson
sunset in background
In one flight, a small piece of duct tape interfered with one of the propellers and was ejected during flight. The separation happened to be caught on video, which shows up clearly after a bit of judicious image enhancement of individual video frames.
Progress is also being made on the second training vehicle, a TIFO, which started initial taxi tests today.
The SuperFlea, TIFO and control systems are on loan from the AIAA Tucson Section.
28 March 2005
Wing passes stress test
The full scale SVC wing passed a structural load test equivalent to a 3 g load case. The test applied to a lightweight version of the vehicle, with minimal avionics and battery weight. Only positive (i.e. noninverted) loading was tested.
Load test using water ballast
Wing bending strength is of particular concern because the wing is so thin. Thickness is 8.5 % and 2.6 % at the tip and root, respectively.
On the other hand, the thinness is offset somewhat by the high taper ratio (4.03) and low aspect ratio (3.42), compared to more conventional airplane wings. The two factors tend to simultaneously reduce the bending moment at the root and increase the amount of structural material available at the root, where it's needed most.
The rationale for the wing design is to reduce labor requirements by constructing the wings from off-the-shelf sheets of styrofoam. The airfoil is a modified flat plate, with minimal shaping at the leading and trailing edge. Structural reinforcement consists of strategically placed strips of fiberglass tape.
In other news, the full scale SVC is being prepared for its first powered flight.
22 March 2005
The team met at the physics lab at Pima West Campus to begin contruction of two slowflyer models for the purposes of flight training, and to familiarize team members with the unique characteristics of tailless airplanes that have swept, low aspect ratio wings. This configuration is similar to the main UAV we'll be flying at the competition.
The vehicles and control systems were generously provided by the AIAA Tucson Section, and were originally intended for the November 2003 Kids Club session that celebrated the Centennial of Flight.
Experiments in night testing
In an effort to make subscale flight testing easier to schedule, we're experimenting with night testing, when wind tends to die down in the local area. The subscale vehicles are sensitive to wind. The first iteration was to add an internal chemical lightstick
Roll angle was difficult to see in flight, which lead to the next iteration. High intensity blue LEDs were installed. These have yet to be flight tested.
2 March 2005
UA Undergrad Research Forum
At the University of Arizona, the 18th Annual Undergraduate Research Forum took place on 16 February. On display were undergrad research projects from many different UA departments, including a a display of a machine vision project by Kate Taralova. We plan on using her software in the IARC competition this summer.
Test images of an actual building at the competition are displayed on the board behind the computer monitors (see below). Also on display is the carrier airplane that is designed to carry out the reconnaissance mission at the competition, as well as carry a secondary drop vehicle (not shown).
Machine vision display
(L to R) Tim Spriggs, Kate Taralova, Frank Manning
University of Arizona photo
University of Arizona photo
24 February 2005
New team member
Welcome to new team member Chris Discenza, who is a UA student.
18 February 2005
First club meeting
Today was the first meeting of the newly-formed Pima UAV Robotics Club.
15 February 2005
SVC upper fuselage built
Most of the upper fuselage structure of the SVC has been built. Outer coverings remain to be applied.
7 January 2005
The IARC application and fee were received. The PCC team is now an official entry in the competition.
9 December 2004
First flight of full scale SVC
First flight of full scale SVC -- controls fixed, unpowered, hand launch. Airframe base only, upper fuselage not installed:
30 November 2004
Construction of full scale SVC begins
Construction of full scale SVC begins 28 November. Foam blanks cut for airframe base, as of 29 Nov:
(Added 2007/3/15) Safety always has priority. Otherwise this is what can happen:
At the AIAA Student Work Facility in Tucson, an accident with a bandsaw occurred during cutting of the SVC foam blanks late in the day on 28 November. In the accident, the saw cut halfway through the bone of the left index finger.
The injury was to a team member who was a nonstudent and therefore should have known better. The team member wrapped his finger in a paper towel, cleaned up most of the blood on the floor, drove himself to the local ER, got stitched up, then came back and finished the cut in the wee hours of 29 November.
17 November 2004
New team member
Welcome to new team member Robert Wagoner, who has extensive UAV experience, and is president of the Electric Jet Factory.
14 November 2004
Added new flight test log. Exercised test sites adajacent to lab facility.
9 November 2004
Added new flight test log:
1 November 2004
Added new flight test log:
Last updated 2008/11/06 FLM