Included below are videos showing some of our test flights, as well as other hardware-related activities as part of this project. We've provided three different file types (with estimates on the file size). Click on the links listed below to either download or watch the videos on-line. Note that there are three different sets of videos: the "small" set (recorded in mpeg1 format) designed for quick downloads or streaming video over the internet (RealPlayer, Windows Media Player and QuickTime should be able to play this file), the "large" set (recorded in mpeg2 format) which should be downloaded before viewing, and a Windows Media Video format. If you are running linux, the latest verision of mplayer should be able to view any of the mpeg videos. If you are using Windows Media Player, you should be able to view the .wmv files without a problem. Otherwise, make sure that you have downloaded the proper codecs and you should be able to view these files. If you have questions or want more info, feel free to contact us using our contact email listed here

January 31st, 2007

ICRA '07 Video Submission

This video submission to the ICRA '07 conference provides an overview of the RAVEN (Real-time Autonomous Vehicle indoor test ENvironment) used in all of the indoor fixed-wing experiments to date. The videos are from test flights which took place before on January 31st, 2007. This video was presented at the ICRA '07 conference, which took place in Rome, Italy on April 10th-14th, 2007.

Video: wmv (71 Mb), mpeg1 (50 Mb)

August 8th, 2006

Multi-Vehicle Search and Track Flight Test - Part 2

The purpose of this flight test was to demonstrate the platform's capability to implement multi-vehicle missions using the task advisor to run the entire mission. In this video we demonstrate the tasking system's capability to command two vehicles to search different areas of the room and track multiple objects - both stationary and moving. In this test, we have three R/C vehicles on the floor, and the two air vehicles search for them. Note that the system does not know where the ground vehicles are going to be a-priori. The videos are from a test flight which took place on August 8th.

Google Video: Shortened Video, Full Length Video

Video Link: wmv (24 Mb) (50 Mb), mpeg1 (50 Mb), mpeg2 (202 Mb)
Shortened: wmv (14 Mb) (32 Mb), mpeg1 (30 Mb), mpeg2 (119 Mb)

August 8th, 2006

MILP Obstacle Avoidance Tests

In this flight test, a Mixed Integer Linear Programming (MILP) based navigation algorithm was used to guide the air vehicle around an obstacle placed in the room. Here, we show two different flight tests. In the first test, a large penalty in the cost function is placed on trajectories which cause the vehicle to fly around the obstacle - hence, forcing the algorithm to calculate a vehicle trajectory over the obstacle. In the second test, a large penality in the cost function is placed on trajectories which cause the vehicle to fly above the obstacle - hence, forcing the algorithm to calculate a vehicle trajectory around the obstacle. The videos are from a test flights which took place on August 8th, 2006.

Flight Test 1: wmv (12 Mb) (25 Mb), mpeg1 (24 Mb), mpeg2 (94 Mb)
Flight Test 2: wmv (10 Mb) (22 Mb), mpeg1 (21 Mb), mpeg2 (82 Mb)

July 31st to August 2nd, 2006

Recharge Landing Tests

The purpose of this flight test was to demonstrate the platform's capability to land an air vehicle and recharge it automatically using the platform's automatic battery recharging system. In these videos we demonstrate the system's capability to land on the recharge platform's landing pad and to start recharging the vehicle's battery after landing. The videos are from a test flights which took place on July 31st to August 2nd, 2006.

Video Part 1: wmv (31 Mb), mpeg1 (29 Mb), mpeg2 (115 Mb)
Video Part 2: wmv (20 Mb), mpeg1 (19 Mb), mpeg2 (72 Mb)

July 31st, 2006

Airplane Tracking Test

The purpose of this flight test was to demonstrate the platform's capability to track vehicles and display them using the platform's operator interface. In this video we demonstrate the system's capability to acquire the position of an airplane flown by a pilot and display the data using the platform's 3D interface. The videos are from a test flight which took place on July 31st.

Video Links: wmv (11 Mb) (25 Mb), mpeg1 (24 Mb), mpeg2 (93 Mb)

May 25th, 2006

Multi-Vehicle Search and Track Flight Test

The purpose of this flight test was to demonstrate the platform's capability to implement multi-vehicle missions. In this video we demonstrate the tasking system's capability to command two vehicles to search different areas of the room and track an object. In this test, we drive a R/C car around on the floor, while two air vehicles search for it. Once a vehicle finds the car, it dives down and starts following it while it remains in its flight area. Then, after the car switches flight areas, the vehicle system commands the other vehicle to track the vehicle, while the original vehicle goes back to searching the region for another object. This test was part of a larger test suite to examine the platforms tasking capability. The videos are from a test flight which took place on May 25th.

Video Links: wmv (4 Mb), mpeg1 (23 Mb), mpeg2 (90 Mb)

March 25th and 29th, 2006

Multi-Vehicle Coordinated Flight Tests

Following our initial Formation Flight Tests in February, we sought to make the vehicle interface to the "outside" world much more simple. Since our ultimate goal is to test the system in long-term missions (where teams of vehicles would be commanded to perform tests), we needed to make the interface so that a tasking system could command the vehicles to perform coordinated flight activities. The following two videos were part of a large test suite where we sought to run tests to show that we could command using a remote program over the network in real-time. In the first video, we show a formation flight test with three vehicles, where we setup a program on a seperate computer which provides the leader with a flight plan. In the second test, we setup a remote program which coordinates the movements of the three vehicles over the network. Note that in this test the vehicles are flying a tilted ferris-wheel like pattern. The videos are from flights which took place on March 25th and March 29th.

Video Links - Part I: wmv (11 Mb), mpeg1 (22 Mb), mpeg2 (87 Mb)
Video Links - Part II: wmv (16 Mb), mpeg1 (32 Mb), mpeg2 (127 Mb)

March 9th to March 22nd, 2006

Vision Sensor Autonomous Landing Tests

Following our use of the vision sensor for positioning of a quad rotor in 2005, the main purposes of these tests was to use the vision sensor to provide accurate vehicle position and attitude information relative to an object. As a result, the sensor could be used to assist a vehicle trying to land on a (moving) object. For this test, first, we started by setting the vision sensor at a known location in the room to make sure that once the vehicle was hovering over the sensor. Note that in all of these tests, the sensor only provided the vehicle with it's x-, y-, and z-position information relative to the location of the sensor. Once the vehicle was hovering over the sensor, the sensor was used to provide the vehicle's x-, y- and z- location. Part I shows a video of this test. Next, the plexiglass table was setup to demonstrate a landing over the vision sensor. Part II shows a video of this test. Next, we built an object to be a "simulated" aircraft carrier which could be driven remotely by computer. Here, we used a mounted a landing platform with the vision sensor target below a chicken wire landing platform on top of a rover which could be driven remotely. In Part III, we show a test where the vehicle lands on the platform when it is not moving. Finally, the rover was commanded to move and the vehicle is provided a general reference location of the vehicle to fly to. Once the vehicle is over the "carrier" it uses the vision sensor to provide it position as it tries to land on the carrier. This flight test was repeated multiple times, and a video of one of these landing tests was taken on March 22nd.

Video Links - Part I: wmv (15 Mb), mpeg1 (32 Mb), mpeg2 (124 Mb)
Video Links - Part II: wmv (14 Mb), mpeg1 (28 Mb), mpeg2 (111 Mb)
Video Links - Part III: wmv (11 Mb), mpeg1 (24 Mb), mpeg2 (92 Mb)
Video Links - Part IV: wmv (14 Mb), mpeg1 (30 Mb), mpeg2 (119 Mb)

February 28th, 2006

Following a Ground Vehicle

Similar to the Formation Flight Tests, we setup the positioning system to acquire the position of a remote control car. Then, the quad rotor was commanded to follow the car as we drove it around the room. Therefore, a person is driving the car around the room and the quadrotor is trying to follow it. This flight was repeated multiple times, and a video of one of the tests was taken of one of the flights on February 28th.

Video Links: wmv (8 Mb), mpeg1 (25 Mb), mpeg2 (100 Mb)

February 26th, 2006

Formation Flight Tests

Two autonomous quadrotor vehicles were setup to fly a series of formation flight tests. In these tests, one vehicle was designated the "leader" and the other was commanded to fly a set distance from it. A series of flights were made throughout the week, and two videos were taken on February 26th to document these tests.

Video Links - Test I: wmv (9 Mb), mpeg1 (18 Mb), mpeg2 (69 Mb)
Video Links - Test II: wmv (8 Mb), mpeg1 (17 Mb), mpeg2 (65 Mb)

January 31st, 2006

Autonomous Multi-Vehicle Flight with Auto Take-off / Landing

Two cordless quadrotor vehicles were given a set of waypoints from take-off to landing. Both vehicles independantly flew trajectories successfully - although there was no coordination between vehicles at this point. This flight was repeated multiple times - a video was taken of one of the flights on February 1st.

Video Links: wmv (10 Mb), mpeg1 (21 Mb), mpeg2 (83 Mb)

January 25th, 2006

Cordless Autonomous Multi-Vehicle Flight

The tether has been removed! At first, we set up the system to fly one vehicle, but it was flying so well that we decided to try to fly two - and it worked. Both vehicles were given a hover waypoint, activated and released near the hover waypoint. Both vehicles independantly flew trajectories successfully for several minutes - no coordination between vehicles at this point. This flight was repeated multiple times - a short video was taken of one of the flights on January 26th.

Video Links: wmv (7 Mb), mpeg1 (15 Mb), mpeg2 (57 Mb)

January 21st, 2006

First Vicon System Single Vehicle Flight Tests

In mid-January 2006, we began using a different indoor positioning system in this platform manufactured by Vicon Peak. In this test, the two cords attached to the vehicle in the video are a power cord (so we didn't have to continually re-charge the vehicle) and an IMU cord (which sent attitude information via RS-232). This flight was made on January 22nd, 2006 and repeated multiple time. A video of this flight was taken of one of the flights on January 22nd.

Video Links: wmv (17 Mb), mpeg1 (22 Mb), mpeg2 (86 Mb)

November 30th, 2005

Flight Summary of Video Sensor Flight Tests (May to Nov 2005)

This video summarizes the initial flights tests from late May to the end of November in 2005. In these tests, we use sensor designed to provide vehicle position and attitude with one camera. A small wireless camera was used to take video of the vision sensor target in all of these tests. Note that the cords attached to the vehicle in most of these tests are a power cord (so we didn't have to continually re-charge the vehicle's batteries during the flight tests) and an IMU cord (which sent attitude information to the off-board computer system via RS-232). This video was originally made on November 30th, 2005 and a portion of this video was re-made on April 23rd, 2006 for release on this website.

Video Links: wmv (10 Mb), mpeg1 (25 Mb)